22 annotated bibliographies AMA style

232

EffEctivEnEss of intuitivE Eating intErvEntion through tExt MEssaging aMong collEgE studEnts

Tessa J. Loughran Illinois State University

Tammy Harpel Illinois State University

Rachel Vollmer Bradley University

Julie Schumacher Illinois State University

This study examined the effects of an intuitive eating (IE) text mes- saging intervention on the IE habits, perceived stress, and perceived self-efficacy of college students in comparison to an electronically emailed handout with the same information. Undergraduate students at a Midwestern university (n=300) completed a pre-intervention survey online which assessed IE practice (Intuitive Eating Scale), perceived stress (Perceived Stress Scale), and self-efficacy (General Self-Effi- cacy Scale and Eating Habits Confidence Survey). Participants were randomly divided into a control (n=150) and intervention (n=150) group. The intervention group received five weeks of intervention with weekly IE texts, and the control received the same IE information in one emailed handout. Following the intervention, all participants com- pleted the post-intervention survey with the same measures. A total of 146 (n = 99 intervention, n = 47 control) participants completed the pre-and post-intervention surveys. Paired t-tests and linear regressions were used for analyses. The results showed the IE texting interven- tion significantly increased total IE habits. Additionally, IE texting was found to increase GSE scores and to limit increases in PSS levels. The results of this study provide evidence that texting can be a successful platform for increasing IE behaviors among college students.

KEYWORDS: Intuitive Eating, Mindful Eating, Nutrition, Perceived Stress, Self-Efficacy, Texting Intervention

Effectiveness Of Intuitive Eating Intervention Through Text Messaging / 233

The prevalence of obesity and overweight among adults in the United States is alarm- ing, with more than one-third of Americans classified as obese and more than two-thirds as overweight (Ogden, Carroll, Kit, & Flegal, 2014). Overweight and obese adults are at high-risk for developing a number of health complications such as type II diabetes, cardio- vascular disease, and some types of cancers (Hruby et al., 2016). A critical time period to prevent overweight may be the college age years given that college students experience newfound food independence, which may result in the formation of poor eating habits, such as overeating, or binge eating, that place them at high risk for obesity (Kelly-Weed- er, Phillips, Leonard, & Veroneau, 2014; Smith-Jackson & Reel, 2012). Additionally, college students with high perceived stress levels are more likely to experience emotion- al eating, which may also lead to weight gain (Wilson, Darling, Fahrenkamp, D’Auria, & Sato, 2015).

The most common form of weight loss intervention used to combat weight gain is calorie restriction; however, this weight loss method is short-term, often resulting in the re- gaining of the lost weight (Mann et al., 2007). Recently, intuitive eating (IE) has gained recognition as a successful weight loss inter- vention method. IE can be defined by three main principles: (a) eating for the purpose of providing the body with energy, not due to emotional cues, (b) listening to the body’s hunger and satiety signals, and (c) uncondi- tional permission to eat (Mathieu, 2009). The use of IE methods with college students has been associated with weight loss, lower blood lipid levels, and improved cardiovascular risk (Hawks, Madanat, Hawks, & Harris, 2005).

Text messaging intervention programs may be an appropriate platform for changing health behaviors among college students be- cause these programs have been described as effortless, but effective, due to the automated

text message reminders (Obermayer, Riley, Asif, & Jean-Mary, 2004). One pilot texting program for MyPlate nutrition education showed that participants had a better un- derstanding of the MyPlate guidelines and increased fruit consumption in comparison to the control group that received a mailed brochure containing the same information (Brown, O’Connor, and Salvaiano, 2014). These results suggest that a text messaging IE nutrition intervention may be an effective, efficient way to reach the college population in order to prevent excessive weight gain.

Currently, to the authors’ knowledge, no studies have evaluated the success and overall effectiveness of using text messaging as a platform for IE intervention to combat obesogenic behaviors among college stu- dents. Compared to traditional intervention methods, reaching college students through a text messaging platform may increase success and adherence to IE guidelines in a relatively cost-effective manner. In turn, increased ad- herence to IE methods may promote healthy eating habits in college students, decreasing the overall risk for excessive weight gain. The current study addresses this gap in the literature, with the intent of informing inter- ventions for the college student population. The results of this study can help determine if IE intervention through text messaging can improve eating habits and, subsequent- ly, provide evidence that text messaging is a successful platform for IE intervention in the college student population.

Specifically, the study investigated the following research questions:

(a) Does intuitive eating intervention through text messaging have a greater influence on the overall intuitive eating habits of college students than an elec- tronically emailed handout with the same information?

234 / College Student Journal

(b) Does intuitive eating intervention through text messaging have a greater influence on college students’ per- ceived self-efficacy in relation to diet than an electronically emailed handout with the same information?

(c) Does intuitive eating intervention through text messaging have a greater effect on college students’ perceived stress than an electronically emailed handout with the same information?

Methodology Participants and Recruitment

After receiving approval from the Uni- versity Institutional Review Board, potential participants were identified and contacted through the University’s listserv of students who had indicated they were willing to re- ceive emails regarding research. In order to participate, individuals must have met the following criteria: (1) a current university student, (2) between 18 to 24 years of age, (3) possess a personal smartphone with the abil- ity to receive standard text messages, and (4) live within a 15-minute walking distance to campus. No exclusion criteria existed for gen- der, race, ethnicity, or income level. The goal sample size was 300 participants, which was randomly divided into the control (n=150) and intervention (n=150) groups.

Procedure From the time the initial recruitment

email was sent, participants had two weeks to volunteer and consent for the study and complete the pre-intervention surveys. A reminder email was sent one week after the initial recruitment email was sent. Informed consent was obtained from each participant using the online platform utilized by the re- searcher’s university. Cell phone numbers and email addresses were obtained from par- ticipants upon their consent for participation.

Following consent, participants completed several questionnaires, including the Intuitive Eating Scale (IES), Perceived Stress Scale (PSS), General Self-Efficacy Scale (GSE), and Eating Habits Confidence Survey, again using the secure online survey system. It was estimated that the online surveys would take each participant approximately 15 to 30 min- utes for completion.

Once the pre-intervention surveys were completed and cell phone numbers and email addresses were obtained from participants, 150 participants were randomly assigned to the texting intervention group, and 150 participants were randomly assigned to the control group. The control group received a general Healthy Eating Behaviors interven- tion email one week after the pre-intervention survey closed. The intervention group began the Healthy Eating Behaviors Text Messaging Program one week after the pre-intervention survey closed.

Intervention program. The texting pro- gram was five weeks long and consisted of a total of 10 text messages, at a rate of two per week (Table 2). Text messages, based on the 10 IE principles, were constructed in 160 characters or less. The text messages were sent to participants through a mass text mes- saging provider, EZ Texting, Monday through Saturday, at either 11:00am or 5:00pm CST. Participants in the intervention group were able to opt out of the study at any time by texting STOP.

The control group received the same messages as the texting intervention group, however, all the messages were delivered at one time as a PDF handout through email. Control group participants were able to reply “STOP” by email at any time to withdraw from the study.

One week after the five-week intervention was completed, both intervention and control participants were sent an email with a link to complete the online post-intervention survey.

Effectiveness Of Intuitive Eating Intervention Through Text Messaging / 235

Table 1. Intuitive Eating Principles Descriptions

IE Principles Description

Principle 1 Reject the diet mentality: Ignore magazine articles and diet books that provide short-term weight loss. Give all of your attention to IE methods.

Principle 2 Honor your hunger: Prevent excessive hunger and rebuild trust with yourself and food.

Principle 3 Make peace with food: Give yourself unconditional permission to eat. Do not deprive yourself in order to prevent uncontrollable cravings and binges.

Principle 4 Challenge the food police: There are no “good” or “bad foods”.

Principle 5 Feel your fullness: Listen to the internal body signals of hunger and fullness.

Principle 6 Discover the satisfaction factor: Eating foods that you like, in a comfortable environment, allow- ing you to feel satisfied.

Principle 7 Cope with your emotions without using food: Resolve emotional issues without using food.

Principle 8 Respect your body: Do not be overly critical or unrealistic of your body type and genetic makeup.

Principle 9 Exercise–feel the difference: Participate in activities that promote exercise that you enjoy.

Principle 10 Honor your health–general nutrition: Choose foods that are good for your health and that you enjoy the taste of.

Table 2. Intuitive Eating Texts

Text Message Description

1 Forget all methods of dieting today! Focus on listening to what your body is telling you.

2 Ask yourself before eating today, is your body telling you it’s hungry? You may be experiencing a behavior or emotion urging you to eat, and no actual hunger.

3 Sometimes our bodies crave sweet treats. Remind yourself that it’s okay to enjoy your favorite foods in moderation.

4 Remember: There are no good or bad foods! Aim for a well-balanced diet.

5 Feel your fullness: Listen to the internal body signals of hunger and fullness.

6 Sit down and enjoy your dinner today. Turn off the television and focus on the meal you’re eating without distractions.

7 Feeling stressed or upset? Eating when you’re not hungry won’t help you manage your emotions.

8 Respect and love yourself. Remember you are different from all others and you should appreciate your genetic makeup!

9 Get out and exercise today! Whether it’s an extra-long walk or perhaps a new fitness class at the gym–your body will thank you for it.

10 Honor your health–choose foods that make you feel good and taste good!

236 / College Student Journal

Upon completion of the post-intervention sur- vey, participants were able to click a link that redirected them to a separate webpage where they could provide their name, email address, and mailing address to enter into a random drawing for one of four $25 gift cards.

Measures The online pre-intervention survey mea-

sured intuitive eating behaviors, perceived stress, general self-efficacy, and self-efficacy related to diet. Additionally, participants an- swered demographic questions (age, gender, year in college, ethnicity, distance live from school, and employment) and reported height and weight (which were used to calculate BMI). The online post-intervention survey contained all of the measures from the pre-in- tervention survey, with the exception of the demographic-related questions that were unlikely to change during the five-week inter- vention period.

The Intuitive Eating Scale—2 (IES) is a 23-item measure that assesses the partici- pants’ eating habits and use of IE (Tylka & Kroon, 2013). The IES has four subscales, which include unconditional permission to eat, eating for physical rather than emotion- al reasons, reliance on hunger and satiety cues, and body-food choice congruence. The questionnaire uses a 5-point Likert scale (1 = strongly disagree, 5 =strongly agree). Each of the four subscales are scored by averaging each subset of questions in each subscale, while the total IES is scored by averaging all items. The average IES scores may range from one to five, with five indicating high intuitive eating practice. The Intuitive Eat- ing Scale has been found to be reliable and valid for the undergraduate college student population (Tylka & Kroon, 2013). Cron- bach’s alpha scores for the pre-intervention survey (.813) and post-intervention survey (.849) showed high levels of reliability for this study.

Perceived stress was measured by the Perceived Stress Scale (PSS) (Cohen, Bruner, Kuryluk, & Whitton, 2015). The Perceived Stress Scale (PSS) is a 10-item questionnaire that evaluates participants’ stress levels in re- lation to their feelings within the past month using a 4-point scale (0 = never, 4 = very often). The items of this scale are summed, with scores ranging from 0 to 40, and a high- er score indicating a higher perceived stress level (Cohen et al., 2015). This scale has been found to have considerable reliability and validity when used within the college popula- tion to assess perceived stress levels (Cohen, Kamarck, & Mermelstein,1983). The scales used for this study showed high levels of reliability based on Cronbach’s alpha scores for the pre-intervention survey (.877) and post-intervention survey (.892).

The Eating Habits Confidence Survey (EHCS) was used to evaluate participants’ confidence in accomplishing a certain be- havior for the next six months (Sallis, Pins- ki, Grossman, Patterson, & Nader, 1988). Ultimately, such confidence helps determine possible behaviors that might occur during di- eting. The scale contains 20 items, with four subscales that assess resisting relapse, reduc- ing calorie intake, reducing salt consumption, and reducing fat consumption (Sallis et al., 1988). Using a 5-point Likert scale (1 = I know I cannot; 5 = I know I can), participants report how confident they feel that they can accomplish a given eating behavior, with a maximum summed score of 100 points. Greater scores indicate higher self-efficacy levels. This scale was previously found to be reliable and valid among adult women (Deck- er, & Dennis, 2013). The Cronbach’s alpha coefficients were adequate for the pre-inter- vention survey (.795) and post-intervention survey (.856).

The General Self Efficacy (GSE) ques- tionnaire was used to assess participants’ general confidence levels in completing

Effectiveness Of Intuitive Eating Intervention Through Text Messaging / 237

or coping with difficult tasks or issues (Schwarzer & Jerusalem, 1995). The scale contains 10 items and uses a 4-point Likert scale (1= not true at all; 4 = exactly true). The scores were summed together ranging from 10 to 40, with higher scores indicating greater self-efficacy. This scale was found to be reliable and valid when measuring gener- al self-efficacy in adult individuals aged 18 to 87 years (De las Cuevas & Peñate, 2015). The results of the Cronbach alpha scores for this survey showed high reliability for the pre-intervention survey (.853) and post-in- tervention survey (.887).

Data Analysis IBM SPSS Statistics Version 24 soft-

ware was used for data analysis. Descriptive statistics were calculated for participant characteristics and variables of interest. To evaluate the effects of the intuitive eating texting intervention on college students’ overall intuitive eating habits, perceived stress, and diet self-efficacy in comparison to the control, paired t-tests were used with a significance value of p<0.05. Control and in- tervention groups’ pre- and post-intervention Intuitive Eating, Perceived Stress, General Self-Efficacy, and Eating Habits Confidence were compared with paired t-tests to assess if the intervention program was associated with significant change in these variables. Additionally, linear regression was used to assess if change in Eating Habits Confidence and Perceived Stress was associated with the intuitive eating texting intervention. In each regression, participant change in BMI, race, gender, and ethnicity were entered as covari- ates and pre-intervention scores were used as control variables in the equation.

Results Participants

A total of 6,035 students, between the ages of 18 to 24 years, were recruited for the study. For the pre-intervention survey, 526 individ- uals opened the survey, 412 consented to par- ticipate, and 300 fully completed the survey. Individuals who completed the survey were randomly assigned to either the intervention or control group. Eight participants in the in- tervention texting group and one in the email control group replied “STOP” and opted out of the study.

After the five-week intervention was com- pleted, 227 participants opened the post-inter- vention survey, and 146 (n = 99 intervention, n = 47 control) fully completed the survey. Of the 146 participants who fully completed both the pre-intervention and post-interven- tion surveys, the majority of participants were 18 years of age (70%), white (90%), female (85%), college freshmen (75%), and currently unemployed (75%).

IES Pre- and Post-Test Score Means Within Groups

Mean IES scores were calculated for both the control and intervention groups. Table 3 shows the pre-intervention and post-interven- tion survey means for both groups, as well as changes between the means. Within the inter- vention group, scores for the IES total score significantly increased by .096, t(98) = 2.564, p <.05. Additionally, for the intervention group, change in one of the four subscales within the IES scale was found to be signif- icant. Specifically, the IES subscale Reliance for Hunger and Satiety Cues significantly in- creased by .201, t(98) = 2.866, p <.005. Both of these significant findings within the inter- vention group show an increase in intuitive eating behavior at post-intervention. Changes within the intervention group for IES scores were significant; however, within the control

238 / College Student Journal

group, there were no significant differences between the pre-intervention and post-inter- vention survey IES scores.

PSS Pre- and Post-Test Score Means Within Groups

Mean PSS scores were calculated for both the control and intervention groups. Table 3 shows the pre-intervention and post-interven- tion survey means for both groups, as well as changes between the means. Scores sig- nificantly increased by 1.303 for PSS scores, t(98) = 2.214, p <.05, among the intervention group, indicating an increase in perceived stress from pre- to post-intervention. While not significant, the PSS score also increased for the control group, with an increase of 1.660, t(46) = 1.926, p>.05. Although per- ceived stress increased in both groups, the control group post-intervention score was higher than the intervention group post-inter- vention score. Therefore, despite the fact that the increase was not significant, the control group had a greater increase in perceived stress than the intervention group.

EHCS and GSE Pre- and Post-Test Score Means Within Groups

Mean GSE and EHCS scores were cal- culated for both the control and intervention groups. Table 3 shows the pre-intervention and post-intervention survey means for both groups, as well as changes between the means. Although the results were not statis- tically significant, the results indicated an increase in both EHCS and GSE scores for the intervention group. Average EHCS scores increased by .333, and GSE scores increased by .534, indicating both perceived self-effica- cy and perceived diet-efficacy increased from pre- to post-intervention. Within the control group, GSE scores decreased by -.987, while EHCS scores increased by 1.617.

Intervention Effects on IE Linear regression was used to assess if the

IE intervention was associated with change in IES scores when controlling for pre-interven- tion scores and other variables. Specifically, participants’ change in BMI, age, gender, year in college, and race were entered as covariates

Table 3. Intuitive Eating Intervention Means and Paired Sample T-Test Results

Pre-Survey Post-Survey Change

Scales INT CNT INT CNT INT CNT

IE Total 3.27 3.24 3.37 3.24 .096* .001

3.01 3.06 2.97 3.05 -.037 -.011

3.21 3.06 3.32 3.10 .106 .040

3.50 3.41 3.70 3.39 .201** -.011

3.52 3.75 3.63 3.70 .114 -.057

EHCS 48.06 52.30 48.39 53.92 .333 1.617

GSE 30.83 32.81 31.36 31.83 .534 -.987

PSS 18.01 17.98 19.31 19.64 1.303* 1.660

Note. N=146, Intervention, N=99, Control N=47. *p<.05, **p<.005. a. Unconditional permission to eat b. Eating for physical rather than emotion reasons c. Reliance on hunger and satiety cues d. Body-Food Choice Congruence

Effectiveness Of Intuitive Eating Intervention Through Text Messaging / 239

in the regression model, with the IE pre-inter- vention survey score used as a control vari- able and IE intervention (texting intervention versus control group membership) serving as the independent variable.

The regression model, excluding the in- dependent variable (IE intervention), was statistically significant, R2 = .528, F(6, 139) = 25.093, p <.0005. Pre-intervention IES sur- vey scores were significantly correlated with post-intervention IES scores. The pre-inter- vention survey scores remained significantly correlated after the independent variable was added to the model; however, the independent variable was not significantly related to IES post-intervention survey scores. See Table 4 for detailed regression results. Therefore, the results of the full regression analysis indicat- ed the IE texting intervention did not relate to post-intervention survey IES scores.

Intervention Effects on Self-Efficacy A linear regression was used to assess if

IE intervention was associated with change in EHCS scores among groups when con- trolling for pre-intervention scores and other variables. Participants’ change in BMI, age, gender, year in college, and race were also entered as covariates in the regression model. The pre-intervention EHCS survey score was used as a control variable.

The results of the regression model, when excluding the independent variable (IE intervention), was statistically significant, R2 = .327, F(6, 139) = 11.234, p <.0005. Pre-intervention EHCS survey scores were significantly correlated with post-intervention EHCS scores. The pre-intervention survey scores remained significantly correlated after the independent variable was added to the model; however, the independent variable was not significantly associated with EHCS post-intervention survey scores. See Table 5 for full details on this regression. Thus, the

Table 4. Intervention Effects on IE

Post-Survey IES Scores

Model 2

Variable B Beta t p

Constant -.589 -.695 .488

IES Pre-Survey .741** .683 11.423 .000

Age .083 .232 1.795 .075

Race -.014 -.019 -.320 .750

Year in School -.100 -.211 -1.610 .110

Gender -.058 -.043 -.720 .473

Change in BMI -.043 -.083 -1.411 .160

IE Intervention .097 .094 1.498 .137

R2 .528

F 22.021**

Change in R2 .008

Change in F 2.243

Note. N=146. *p<.05, **p<.001.

Table 5. Intervention Effects on EHCS

Post-Survey EHCS Scores

Model 2

Variable B Beta t p

Constant 11.450 .668 .505

IES Pre-Survey .571** .500 7.040 .000

Age .661 .109 .711 .478

Race -1.892* -.153 -2.154 .033

Year in School -1.530 -.189 -1.208 .229

Gender 1.554 .067 .952 .343

Change in BMI -.433 -.050 -.707 .481

IE Intervention -1.528 -.087 -1.159 .248

R2 .333

F 9.844**

Change in R2 2.99

Change in F 1.343

Note. N=146. *p<.05, **p<.001.

240 / College Student Journal

results of the full regression analysis indicat- ed IE texting intervention did not correlate with post-intervention survey EHCS scores.

Intervention Effects on Perceived Stress Finally, linear regression was used to as-

sess if IE intervention was associated with change in PSS scores among groups. Par- ticipant change in BMI, age, gender, year in college, and race were entered as covariates in the model, with the pre-intervention PSS sur- vey entered as a control variable, and IE inter- vention serving as the independent variable.

The results of the regression model, with- out the independent variable included, were statistically significant, R2 =.414, F(6, 139) = 16.394, p <.0005. Pre-intervention PSS sur- vey scores were significantly associated with post-intervention PSS scores. The pre-inter- vention survey scores remained significantly associated after the independent variable was added to the model; however, the indepen- dent variable did not result in a significant

relationship between PSS post-intervention survey scores. See Table 6 for full details on this regression. Thus, the results of the full regression analysis indicated the IE texting intervention was not associated with post-in- tervention survey PSS scores.

Discussion This study examined the effectiveness of

an intuitive eating texting program in compar- ison to an emailed handout among a college student population. The intervention effects on IE practice, eating habit confidence, and perceived stress levels were examined in the study. The results of the study showed sig- nificant increases in total IES scores within the intervention group, with greater changes for the intervention group than the control group. These findings suggest that IE texting intervention over a five-week period may be more effective in improving IE practice than receiving an emailed IE handout containing the same information. A similar study also found increases in targeted nutrition-related behaviors and knowledge when evaluating the effectiveness of a nutrition education based texting program, Mobile MyPlate (Brown et al., 2014). Brown et al. found that once the Mobile MyPlate intervention was complete, participants showed increases in fruit con- sumption and MyPlate nutrition knowledge when compared to the control group that received a mailed handout with the same in- formation. The results of the Mobile MyPlate study (Brown et al., 2014) and the current study both suggest texting intervention pro- grams can increase desired nutrition-related behaviors more than a handout delivered at a single time.

In addition to the significant increase in total IES scores, the IES subscale Reliance on Hunger and Satiety Cue scores significantly increased among the intervention group. In- terestingly, this subscale actually decreased in the control group that received the one-time

Table 6. Intervention Effects on PSS

Post-Survey PSS Scores

Model 2

Variable B Beta t p

Constant 27.060 -1.974 .050

IES Pre-Survey .646** .602 -9.022 .000

Age -1.157 -.227 -1.571 .118

Race .163 .016 .237 .813

Year in School 1.328 .195 1.336 .184

Gender 2.261 .117 1.764 .080

Change in BMI -.321 -.044 -.666 .507

IE Intervention -.410 -.028 -.396 .692

R2 .415

F 13.989

Change in R2 .001

Change in F .157

Note. N=146. *p<.05, **p<.001.

Effectiveness Of Intuitive Eating Intervention Through Text Messaging / 241

email version of the IE information. These findings suggest that texting IE reminders are a more effective way to remind individuals of their hunger and satiety cues versus reading the information within an emailed handout. In particular, this finding may be explained by the timeliness of the texts sent in the intervention. The texts for this study were purposely sent slightly before meal times, with the intention of reminding participants to focus on their IE skills before consuming meals. On the other hand, at mealtimes, those in the control group would have to remind themselves of what they had read about IE earlier in the emailed handout. Having texts sent before meal times may have been more successful at increasing IE behavior than the emailed handout because participants did not have to recall information or motivate themselves to reference the hand- out at meal times. This finding is consistent with the findings of Obermayer et al. (2004) and colleagues, which showed a texting smok- ing cessation program was successful due to the fact that participants did not have to seek information (i.e., open an email, log into a website). Taken together, the results of our study and other research, suggest that texting programs may increase desired behaviors due to the timing of the text and the minimal effort required to seek or recall information that may promote the targeted behavior.

The current findings showed the interven- tion group reported significantly higher PSS scores in the post-intervention survey. The results for the control group’s average PSS scores were not significant. This suggests an IE texting intervention may have a greater im- pact on helping one to manage their perceived stress level than information that is delivered one time via email. Yet, both intervention and control PSS scores increased from pre- to post-intervention. Perhaps, the increases in scores among both groups can be attributed to the university environment of the population. The majority of the students who participated

in this study were freshman (75%). Therefore, this study took place during the first semester of college for many of the participants, which can be a very stressful time for students, due to the adjustment of being away from home and course workload. Students are also expe- riencing newfound food independence, which can create poor eating habits, such as over- eating, or binge eating (Kelly-Weeder et al., 2014; Smith-Jackson & Reel, 2012). College students with high perceived stress levels are more likely to experience emotional eating (Wilson et al., 2015). In turn, emotional eat- ing and poor eating habits due to the new col- lege setting may be related to high levels of perceived stress, possibly altering PSS scores. Our results may further be explained by a study that examined the relationship between IE and perceived stress among Finnish obese adults (Järvelä-Reijonen et. al, 2016), with the results showing increased levels of perceived stress to be correlated with low IE practice. In terms of our study, the IE texting intervention had a slight impact on PSS scores among the intervention group; however, the IE texting intervention may have had a larger impact in a less stressful context.

In contrast to previous research, the results involving the GSE scores and EHCS scores among both groups were not significant. In particular, Annessi and Gorjala (2010) found that following the implementation of a 6-month intervention program involving nutrition and exercise, increases in self-efficacy were re- lated to increases in desired health behaviors and outcomes among an adult population. Our study, on the other hand, did not provide evidence that IE intervention correlated with increased self-efficacy. It is possible that the length of our intervention played a role in the discrepancy between our findings and those of Annessi and Garjala. Perhaps our IE texting program would have resulted in greater in- creases in self-efficacy if the intervention pro- gram was longer-term, such as the Annessi and

242 / College Student Journal

Gorjala intervention, versus only five weeks in duration. In order to further examine this re- lationship, future research would benefit from utilizing longer interventions.

Although the results were not statistically significant for GSE, it is interesting that the GSE scores among the control group de- creased, while the intervention group GSE scores increased. A study by Moeini and col- leagues (2008) may provide some explanation for these findings. The researchers examined the relationship between PSS and GSE lev- els in Iranian male high school students and found that increased amounts of PSS led to de- creased amounts of GSE (Moeini et al. 2008). In our study, the intervention group reported a smaller increase in PSS when compared to the control group. Given that PSS levels showed a greater increase among the control group when compared to intervention group scores, this may explain why GSE scores decreased in the control group and increased in the in- tervention group.

While this study provides evidence to support the use of IE texting interventions, the study is not without limitations. First, it is unknown if the participants opened the text messages that contained the IE intervention material and, if so, how they used the texts. Based on how the participants handled the text messages, it may be that the intervention group did not receive the full benefits of the texts. In support of this, other studies have successfully included interaction via text be- tween participants and researchers (de Niet et al., 2012; Bauer, de Niet, Timman, & Kordy, 2010). Such interaction between researchers and participants may help monitor and en- courage participant engagement and may, subsequently, increase intervention effec- tiveness. Future research would benefit from including a mechanism that motivates partici- pants to open and read the text messages.

Participant attrition was another issue in this study. A portion of participants who

completed the pre-intervention survey did not complete the post-intervention survey, resulting in more participants within the in- tervention group than in the email (control) group. The variance between group sizes may have produced unbalanced results. The higher attrition of the control group from pre- to post-intervention may, at least in part, be due to the intervention group receiving a reminder text to complete the survey, while the control received an email reminder. Oth- er studies have found the automated text re- minder to be a better form of communication than email or mail for the college-aged, likely because texts do not require self-motivation to seek information by opening an email or logging onto a website (Brown et al., 2014; Hebden et al., 2014; Obermayer et al., 2004). Therefore, the likelihood of attrition may be decreased in future studies through the use of texting reminders for both intervention and control groups.

Additionally, all data in this study were self-reported by participants. Questions re- lated to food behaviors, stress levels, and weight may have caused discomfort for some participants, and participants may not have answered the questions honestly. Finally, given that participants were not screened for pre-existing knowledge of IE prior to this study, it is unknown if participants were ever exposed to IE or had knowledge of IE prior to this study. Pre-existing knowledge and use of IE prior to the study may have affected the degree of change in scores.

Conclusion Previous research suggested texting pro-

grams are effective for delivery of nutrition related information; however, this study was the first to implement IE education in a texting platform. The results of this study found an IE texting intervention significantly increased total intuitive eating habits within a college student population. In addition, our

Effectiveness Of Intuitive Eating Intervention Through Text Messaging / 243

findings also indicated that while all college students in the sample reported greater stress from pre- to-post intervention, the texting intervention was associated with a smaller increase in stress than what was observed in the control group. Therefore, the results of the study provide evidence that texting can be a successful platform for increasing IE behaviors among college students, while also suggesting that a texting intervention may play a moderating role in the perceived stress of college students. Based on the results of other research, we believe texting programs of longer duration would likely produce even greater benefits among the college population than what was observed in our study. Over- all, we believe our results suggest that the implementation of IE eating texting programs on college campuses can increase student healthy eat habits, which may reduce obesity and promote healthy lifestyles.

References Annesi, J., & Gorjala, S. (2010). Relations of self-regula-

tion and self-efficacy for exercise and eating and BMI change: A field investiga-

tion. Biopsychosocial Medicine, 4(1), 10. doi: 10.1186/1751-0759-4-10

Bauer, S., de Niet, J., Timman, R., & Kordy, H. (2010). Enhancement of care through self-monitoring and tailored feedback via text messaging and their use in the treatment of childhood overweight. Patient Edu- cation and Counseling, 79, 315-319. doi: 10.1016/j. pec.2010.03.014

Brown, O. N., O’Connor, L., & Savaiano, D. (2014). Mo- bile MyPlate: A pilot study using text messaging to provide nutrition education and promote better dietary choices in college students. Journal of American College Health, 62(5), 320-227. doi: 10.1080/07448481.2014.899233.

Cohen, S., Bruner, M. R., Kuryluk, A. D., & Whitton, S. W. (2015). Perceived Stress Scale—14. Journal of American College Health, 63(2), 98-108. doi: 10.1517/13543784.12.11.1757

Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of perceived stress. Journal of Health and Social Behavior, 24(4), 385-396. doi:10.2307/2136404

De las Cuevas, C., & Peñate, W. (2015). Validation of the General Self-Efficacy Scale in psychiatric outpa- tient care. Psicothema, 27(4), 410-415. doi:10.7334/ psicothema2015.56

de Niet, J., Timman, R., Bauer, S., van den Akker, E., de Klerk, C., Kordy, H., Passchier, J. (2012). Short message service reduces dropout in childhood obe- sity treatment: A randomized controlled trial. Health Psychology, 31(6), 797-805. doi: 10.1037/a0027498

Decker, J. W., & Dennis, K. E. (2013). The Eating Habits Confidence Survey: Reliability and validity in overweight and obese postmenopausal wom- en. Journal of Nursing Measurement, 21(1), 110-119. doi:10.1891/1061-3749.21.1.110

Hawks, S., Madanat, H., Hawks, J., & Harris, A. (2005). The relationship between intuitive eating and health indicators among college women. American Journal of Health Education, 36(6), 331-336. doi:10.1080/19 325037.2005.10608206

Hebden, L., Cook, A., Ploeg, H. P., King, L., Bauman, A., & Allman-Farinelli, M. (2014). A mobile health intervention for weight management among young adults: A pilot randomized controlled trial. Journal of Human Nutrition & Dietetics, 27(4), 322-332. doi:10.1111/jhn.12155

244 / College Student Journal

Hruby, A., Manson, J., Lu, Q., Malik, V., Rimm, E., Qi, S., & Hu, F. (2016). Determinants and Consequenc- es of Obesity. American Journal of Public Health, 106(9), 1656-1662. doi:10.2105/AJPH.2016.303326

Järvelä-Reijonen, E., Karhunen, L., Sairanen, E., Ran- tala, S., Laitinen, J., Puttonen, S., & Kolehmainen, M. (2016). High perceived stress is associated with unfavorable eating behavior in overweight and obese Finns of working age. Appetite, 103, 249-258. doi:10.1016/j.appet.2016.04.023

Kelly-Weeder, S., Phillips, K., Leonard, K., & Veroneau, M. (2014). Binge eating and weight loss behaviors of overweight and obese college students. Journal of the American Association of Nurse Practitioners, 26(8), 445-451. doi:10.1002/2327-6924.12070

Mann, T., Tomiyama, A. J., Westling, E., Lew, A., Sam- uels, B., & Chatman, J. (2007). Medicare’s search for effective obesity treatments: Diets are not the answer. American Psychologist, 62(3), 220-233. doi: 10.1037/0003-066X.62.3.220

Mathieu, J. (2009). Practice application: What should you know about mindful and intuitive eating? Jour- nal of the American Dietetic Association, 109(12), 1982-1987. doi: 10.1016/j.jada.2009.10.023

Moeini, B., Shafii, F., Hidarnia, A., Babaii, G. R., Bi- rashk, B., & Allahverdipour, H. (2008). Perceived stress, self-efficacy and its relations to psycholog- ical well-being status in Iranian male high school students Social Behavior & Personality: An Inter- national Journal, 36(2), 257-266. doi: 10.2224/ sbp.2008.36.2.257

Obermayer, J., Riley, W., Asif, O., & Jean-Mary, J. (2004). College smoking-cessation using cell phone text messaging. Journal of American College Health, 53(2), 71-78. doi: 10.3200/JACH.53.2.71-78

Ogden, C., Carroll, M., Kit, B., & Flegal, K. (2014). Prevalence of childhood and adult obesity in the United States, 2011-2012. Jama, 311(8), 806-814. doi:10.1001/jama.2014.732

Sallis, J., Pinski, R., Grossman, R., Patterson,T., & Nad- er, P. (1988). The development of self-efficacy scales for health-related diet and exercise behaviors. Health Education Research, 3, 283-292. doi: 10.1093/ her/3.3.283

Schwarzer, R., & Jerusalem, M. (1995). Generalized Self-Efficacy scale. In J. Weinman, S. Wright, & M. Johnston, Measures in health psychology: A user’s portfolio. Causal and control beliefs (pp. 35-37). Windsor, England: NFER-NELSON.

Smith-Jackson, T., & Reel, J. J. (2012). Freshmen women and the “freshman 15”: Perspectives on prevalence and causes of college weight gain. Jour- nal of American College Health, 60(1), 14-20. doi: 10.1080/07448481.2011.555931

Tylka, T. L., & Kroon, V. D. (2013). Intuitive eating scale—2. Journal of Counseling Psychology, 60(1), 137-153. doi:10.1037/a0030893

Wilson, S., Darling, K., Fahrenkamp, A., D’Auria, A., & Sato, A. (2015). Predictors of emotional eating during adolescents’ transition to college: does body mass index moderate the association between stress and emotional eating? Journal of American College Health, 63(3), 163-170. doi:10.1080/07448481.2014 .1003374

Copyright of College Student Journal is the property of Project Innovation, Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.