S68 Poster Abstracts

P58 (continued) Results: There were significant (P 0.01) increases in the

intake of calcium- and vitamin D-rich foods (increased by
discussion, was implemented with a group of 25 Head Start 4 weekly servings), the use of calcium- or vitamin
Center supervisors and staff in San Luis Obispo County, D-containing supplements (8 to 9 percentage-point in-
CA. creases), and following 7 to 8 of the 8 fall preventive
Outcome Measures and Analysis: The program was eval- behaviors (pre: 35% vs. post: 51%), and decreases in overall
uated using a self-efficacy pre- and post-test of the training MORR ( 5-6 risk factors: pre: 19% vs. post: 9%).
participants. Paired t-tests were used to measure change in Conclusions and Implications: This evaluation provides
self-efficacy, skills and outcome expectancies from pre- to evidence that this intervention improves several preventive
post-test. behaviors that are important for decreasing the risk of falls
Results: Major results showed a significant (p 0.005) and fractures in older adults. This project was funded by the
increase in self-efficacy to: 1) communicate nutrition con- Georgia Division of Aging Services, the Northeast Georgia
cepts effectively; 2) influence changes in nutrition related Area Agency on Aging, and the University of Georgia.
attitudes; and 3) find resources to answer questions.
Conclusions and Implications: These results suggest that
facilitated group discussion, led by a skilled facilitator, can P60 Focus Groups to Inform the
be effective in increasing the self-efficacy of paraprofession-
Development of Online Theory Based
als working as nutrition educators.
Education in Breastfeeding Initiation Best
Practices
Jennifer Weddig, PhDc, RD, CLC, Metropolitan State
P59 Fall and Fracture Preventive Behaviors College of Denver, CB 33, Denver, CO 80217;
Increased Following a Community-Based Susan Baker, EdD, Department of Food Science and
Intervention in Older Adults in Georgia Human Nutrition, 214E Gifford Building, Colorado State
Senior Centers University, Fort Collins, CO 80523,
Jennifer Teems, BS, University of Georgia, 871 Meigs sbaker@cahs.colostate.edu; Garry Auld, PhD, RD,
Street, Athens, GA 30606, JET0802@uga.edu; Department of Food Science and Human Nutrition, 229
Tiffany Lommel, MS, RD, LD, University of Georgia, Gifford Building, Colorado State University, Fort Collins,
tsellers@uga.edu; Joan G. Fischer, PhD, RD, LD, CO 80523, auld@cahs.colostate.edu
University of Georgia, jfischer@fcs.uga.edu;
Objective: To determine what factors affect a registered
Jung Sun Lee, PhD, RD, LD, University of Georgia,
nurses ability to support their patient in breastfeeding
leejs@fcs.uga.edu; Sudha Reddy, MS, RD, LD, Georgia
initiation, if registered nurses have access to computers and
Department of Human Resources, Division of Aging
the computer skills necessary to complete an online course,
Services, sureddy@dhr.state.ga.us;
and what hospital policies and procedures, if any, impede a
Mary Ann Johnson, PhD, University of Georgia,
nurses ability to put their breastfeeding knowledge into
mjohnson@fcs.uga.edu
practice with their patients.
Design, Setting and Participants: Focus groups were con-
Objective: The purpose of this study was to evaluate a
ducted at 8 Colorado hospitals with over 40 nurses from
community-based intervention to reduce risk factors related
Labor/Delivery, NICU, Postpartum, and Pediatrics.
to falls and fractures in older adults.
Outcome Measures and Analysis: The focus group tran-
Design, Setting and Participants: The intervention took scripts were analyzed by hand and the main themes of the
place in Georgia senior centers and participants were a focus groups were extracted.
convenience sample (N 849; mean age 74; 83% Results: Most nurses report that: 1. They do not have
female; 55% white, 44% black, 1% other). official protocols for feeding. 2. They are supportive of
Intervention: The 4-month intervention consisted of 16 breastfeeding but lacked the hands on skills necessary to
sessions, eight of which focused on prevention of falls and assist dyads. 3. The infant is not given uninterrupted skin to
fractures, and incorporated components of the Health Be- skin contact during the first 2 hours after birth. 4. They did
lief Model. All 16 sessions included physical activity. not observe a dyad breastfeeding every shift and often never
Outcome Measures and Analysis: Pre- and post-tests were observed an actual feeding during a shift. 5. They are
interviewer-administered and evaluated fall preventive interested in additional education that would allow them to
home safety behaviors, intakes of calcium- and vitamin assess a dyad and provide help for common problems 6.
D-containing foods and supplements, and 6 modifiable They would benefit from an online course with a skills lab
osteoporosis-related risk (MORR) factors (low intake of and that they would have access to high speed internet to
calcium-rich food, non-use of a calcium supplement, non- allow for the course to include streaming video and other
use of a vitamin D-containing supplement, low physical
activity, high risk of falling, and current tobacco use). Continued on page S69
 Journal of Nutrition Education and Behavior Volume 40, Number 4, July/August 2008 Supplement S69

P60 (continued) home and about one-third were experienced as a snack.

Saturated fat and sweets intakes were highest in indulgent
interactive tools 7. That getting CEUs for completion of escape occasions.
the course would be valuable and a motivator for partici- Conclusions and Implications: Attitude and eating occa-
pation. sion need state affect eating behaviors and should be con-
Conclusions and Implications: The focus groups informed sidered when developing tailored interventions to improve
the development of an online course teaching breastfeeding intake and control weight. This project was funded by NIH.
initiation best practices and policies that was conducted
during November 2007.

P62 Improving the Accuracy of a Self-

P61 Segmentation of Midlife Women by Administered Mailed 24-Hour Food Record
Noriko Sudo, PhD, RD, University of Minnesota,
Attitude and Eating Occasion Need State and
Department of Food Science and Nutrition, 1334 Eckles
Relationship to Obesity Indicators, Food Avenue, 225 FScN, St. Paul, MN 55108,
Group and Nutrient Intake sudox008@umn.edu; Courtney Perry, MS, RD, University
Noriko Sudo, PhD, RD, University of Minnesota, of Minnesota, Department of Food Science and
Department of Food Science and Nutrition, 1334 Eckles Nutrition, perry244@umn.edu; Marla Reicks, PhD, RD,
Avenue, 225 FScN, St. Paul, MN 55108, University of Minnesota, Department of Food Science
sudox008@umn.edu; Dennis Degeneffe, MBA, University and Nutrition, mreicks@umn.edu
of Minnesota, Department of Applied Economics, 317
Classroom Office Building, 1994 Buford Avenue, St. Objective: To determine the accuracy of a self-administered
Paul, MN 55108-6040, ddegenef@umn.edu;
mailed 24-hour food record (FR) by comparing the uncor-
Houa Vue, MS, RD, University of Minnesota,
rected FR (as received) to the FR corrected by telephone
Department of Food Science and Nutrition, 1334 Eckles
interview.
Avenue, 225 FScN, St. Paul, MN 55108,
Design, Setting and Participants: Forty midlife women (49
vuex0067@umn.edu; Marla Reicks, PhD, RD, University
of Minnesota, Department of Food Science and 1 5 years) were recruited from flyers on a university campus.
Nutrition, 1334 Eckles Avenue, 225 FScN, St. Paul, MN Women were mailed a FR booklet and a 16-page actual-size
55108, mreicks@umn.edu 2D food model booklet. The FR booklet instructed women
to describe foods consumed and preparation methods/
Objective: To segment women by attitude and eating oc- recipes. Women returned the completed uncorrected FR
casion need state and to determine the relationship be- form by mail and were interviewed by phone to review
tween these segments and obesity indicators, food group missing detail and clarify amounts and omissions (corrected
and nutrient intake. FR).
Design, Setting and Participants: Cross-sectional study Outcome Measures and Analysis: Pearsons correlation
with 200 midlife women (46 1 6 years). coefficients and mean differences were calculated for intake
Outcome Measures and Analysis: Women completed an of energy and nutrients between the uncorrected and cor-
Attitude/Control Belief Questionnaire once and an Eating rected FR.
Occasion Questionnaire for 3 eating occasions during Results: All intakes based on the uncorrected and cor-
3-days when diet records were collected. Cluster analyses rected FR were significantly correlated. Mean difference (1
segmented women based on attitudes toward food and SE) was the highest for energy (154 1 62 kcal) and the
eating occasions based on motivations underlying food lowest for vitamin E (0.57 1 0.40 mg). For all nutrients and
choices. energy, the uncorrected FR overestimated intakes. The
Results: Five attitude segments and six eating occasion most frequent missing data in the uncorrected FR was
need state segments were identified. Estimated marginal inadequate description of foods and inadequate descrip-
means of body mass index, waist circumference, and per- tion of serving sizes. The median number of food items (25
cent body fat were highest in attitude segments defined as and 75 percentiles) per FR with these types of missing data
impulsive eaters. Busy cooking avoiders had a higher was 3.5 (2.0, 5.8) and 1.0 (0.0, 1.0), respectively. Education
energy intake compared to women in other attitude seg-
level may have contributed to more frequent missing data
ments. Energy, total fat, and cholesterol consumption per
in the uncorrected FR.
occasion were highest in routine family meal occasions of
Conclusions and Implications: Even with use of well-
which more than 60% were dinner and eaten at home with
designed booklets, accuracy of mailed FR could be im-
children. The percentage of eating occasions in which
fruits/vegetables were eaten was also highest in routine proved via telephone interviews.
family meal followed by healthy regimen occasions. More
than half of indulgent escape occasions occurred away from