The Effects of Whey and Soy Protein Supplementation on Appetite and Dietary Quality in Overweight and Obese College-Aged Individuals

Background: Over one-third of the United States population is obese. Obesity is a complicated disorder associated with many chronic diseases such as coronary artery disease, diabetes, hypertension and stroke. Many college students are overweight or obese, which may be due to lack of physical activity and unhealthy diets. Both dietary quality and satiety are important factors that may help modify obesity and associated health conditions. The impact of various interventions on these factors has not been clarified. Protein supplementation may be able to improve satiety, but research regarding improvement in diet quality in relation to this supplementation is limited. Objective: To determine the impact of an 8-week protein supplementation intervention on dietary quality and appetite by comparing groups supplemented with whey or soy protein with each other and with an assessment only control group. Methods: In a randomized, control trial with pre-post testing, subjects were randomized to one of three groups, whey protein, soy protein or non-treatment control. Experimental group subjects participated in 8 weeks of protein supplementation, and the control group received no treatment. Dietary quality and appetite were assessed at week 0 and week 8. The primary dietary quality outcome was total Alternate Healthy Eating Index (AHEI) score. The primary appetite outcomes were lab-assessed fasting hunger and satiety. Subjects assigned to the supplement groups were asked to ingest either the whey or the soy protein supplement, providing 21.5 grams of protein per day, on a daily basis for 8-weeks. All laboratory visits were conducted at the University of Rhode Island. Analysis of variance was used to compare within-group and between-group differences in dietary quality and satiety for pre and post measurements. Participants: Nine overweight and obese students (three per group) were recruited from the University of Rhode Island and surrounding areas through classroom announcements, fliers and mass emails. Results: There were no significant differences between the groups at baseline, although, based on the visual analog scales, the control group tended to have higher fasting hunger levels (control: 54.1±27.7mm, experimental: 35.5±26.1mm) and lower fasting satiety levels (control: 35.5±19.3mm, experimental: 56.0±26.5mm). The control group also had lower energy intake (control: 2428.4±1266.9kcal, experimental: 2838.3±1182.5kcal) and lower protein intake (control: 77.3±13.3g, experimental: 99.3±24.4g) based on three day food diaries, as well as higher diet quality (control: 44.0±6.6, experimental: 35.7±6.5) based on the total AHEI scores. No significant time by group or within group differences were found for AHEI scores or visual analog scales for hunger, satiety or appetite. During supplementation, the experimental groups consumed significantly more protein than the control group (experimental: 115.3±25.5 grams, control: 73.7±10.6grams, p=.033). Measures of satiety and hunger were not significantly impacted by the intervention. Conclusion: The addition of a protein supplement to the diet of overweight and obese young adults, with dietary counseling incorporated, did not improve diet quality or suppress fasting appetite. With a larger sample size, the effectiveness of this intervention may be measureable. Overall, this research gathered valuable information for use in interventions in the future.


INTRODUCTION
Obesity has been on the rise in the United States since the late 1980s . In 2008, over 74% of the adult population in the United States was overweight, obese or extremely obese . Obesity is a complicated disorder associated with many chronic diseases such as coronary artery disease, type II diabetes, hypertension, dyslipidemia, stroke, sleep apnea, and some cancers . According to the National College Health Risk Behavior survey, 35% of college students are overweight or obese, which may be due to lack of physical activity and unhealthy diets . College students are within the 18-29 year old population, which is where the greatest increase in weight gain and obesity has been observed . The college environment can be conducive to overconsumption of energy dense food . Levitsky et al. measured weight changes from high school to the first few months of college and reported that between 58 and 71% of the total variance of weight gain was due to increased food consumption and unhealthy eating behaviors . Obesity is partially due to chronic positive energy balance, related to a diet lacking in balance and moderation (U.S. Department of Health and Human Services, 2010)

Diet Quality & Protein
Poor diet quality, as defined by the Healthy Eating Index (Guo, Warden, Paeratekul & Bray, 2004), is associated with obesity and overweight status . Improving diet quality could influence weight status in some individuals. High dietary quality as measured by the Alternate Healthy Eating Index (AHEI), has been correlated with lower risks for major chronic diseases . The AHEI targets food choices and macronutrients associated with reduced chronic disease . The AHEI may be useful as a guideline for reducing the risk of chronic disease .
When compared to other instruments used to measure dietary quality, the AHEI appears to more sensitive.  found that, compared to the original Healthy Eating Index, the AHEI was more closely related to chronic disease risk. These researchers found that the AHEI was almost twice as predictive of chronic disease risk as the original HEI .Little or no research has been conducted on the effects of protein supplementation on diet quality.

Satiety & Protein
Protein is known to be the most satiating macronutrient , which makes it an effective tool for weight loss and management . Higher protein diets have been shown to decrease appetite, as protein increases satiety and metabolic rate . Most studies measuring the effects of protein supplementation on appetite focus on healthy individuals, not the overweight and obese population. Many of these studies look at the postprandial effects of the supplementation following a laboratory meal and not during a free-living situation . This study will measure a free-living setting. In addition, different protein sources affect the body differently in regards to satiety . In a study conducted by DeCassia et al., researchers found that soy protein increased thermogenesis, whey protein decreased respiratory quotient and casein decreased energy intake. That study only included normal weight subjects for the four 7-day experimental, crossover sessions . Some protein supplementation research has been completed in the college population, but the majority of the work looks at the effect on physical performance and examined few nutrition parameters. Soy protein supplementation has not been addressed in the college population in regards to improving dietary quality or satiety. Both whey and soy proteins have been shown to aid in weight loss, lower blood pressure, improve lipid profile, and reduce overall risk of cardiovascular disease in some populations , but have not been systematically compared for their effects on satiety in overweight and obese young adults.

Study Design
This research was a randomized, experimental design conducted over an 8-week period examining the relationships among protein supplementation, dietary quality and appetite. Subjects were randomized to one of three groups, whey protein, soy protein or non-treatment control. Researchers assessed dietary quality and satiety before and after the intervention for all three groups. Subjects assigned to the supplement groups ingested either whey or soy protein supplements which provided 21.5 grams of protein per day, on a daily basis for the 8-week period. The researchers provided the supplements. All laboratory visits were conducted at the Human Performance Laboratory at the University of Rhode Island.

Participants
Nine college-aged men and women between the ages of 18 and 25 were recruited for the study. Subjects were recruited through flyers and classroom announcements at the University of Rhode Island-Kingston campus. This study was part of a larger study where eligibility criteria included having at least two risk factors for cardiovascular disease. Risk factors included waist circumference greater than 40 inches for men and 35 inches for women, body mass index greater than 30 kg / m 2 , sedentary lifestyle, or recent diagnosis of metabolic syndrome, hypertension or dyslipidemia by a physician.
Recent diagnosis of metabolic syndrome, hypertension and dyslipidemia was determined using a medical history questionnaire (See Appendix B). As described later, waist circumference was measured to the nearest 0.2cm using a non-stretchable tape measure at the level of the umbilicus upon exhalation, and body mass index was calculated using the following formula: body weight in kilograms / (height in meters) 2 .
Sedentary lifestyle was defined as participating in less than 30 minutes of moderate intensity physical activity less than three times per week. Exclusion criteria included pregnancy/lactation and vegetarians who consume mainly soy protein. All subjects were required to read and sign two informed consent forms (See Appendix I), approved by URI's Institutional Review Board, before participation in the study.
Participants kept one copy of the informed consent form, while the other copy stayed in their participant folder.

Procedures
Subjects were asked to make three visits to the lab throughout the study, a preliminary

Three-day diet records
Three day diet records were used to record subject food intake for the three days before the intervention began as well as the three final days of protein supplement intake. The form had space for subjects to record a description of the food/beverage consumed, the amount, the total calories and the time as well as any comments they had. Each participant was given instructions regarding estimating portion sizes and locating pertinent dietary information from food labels. The diet records were entered into Food Processor SQL (FP-SQL) for analysis. The three days were averaged for analysis. All data entry was double-checked for accuracy.

Dietary Quality
The AHEI was used to measure the dietary quality after data were entered into FP-SQL. The AHEI is based on 9 items, including vegetables, fruits, nut and soy, ratio of white to red meat, cereal fiber, trans fat, ratio of polyunsaturated fat to saturated fat, duration of multivitamin use and alcohol. Servings of vegetables, fruits, nuts & soy protein and alcohol were determined using the completed diet records. Ratio of white to red meat, cereal fiber, trans fat, ratio of polyunsaturated to saturated fat were determined after food logs were entered into FP-SQL. Duration of multivitamin use was determined by the nutrition and medical history questionnaires. The AHEI has a minimum score of 2.5 and a maximum score of 87.5 . All items, except multivitamin use, were scored from 0-10, with 10 indicating the recommendation was met. Rationale for points system is described in detail in previous studies . Points were awarded for long-term multivitamin use, 2.5 points indicating regular intake less than 5 years and 7.5 for long term use, greater than 5 years. The protein supplements were not included in the total AHEI scores, since the soy protein group would be favored by the addition of one serving of soy protein to each day they recorded during the supplementation period.

Visual Analog Scales (VAS)
Validated VAS were used to measure levels of satiety . Standard procedures were followed in both the laboratory and free-living state . The lab VAS was administered once at the pre-and once at the post-intervention visits to assess fasting appetite perceptions under standardized conditions. Free-living VAS were completed in conjunction with the 3-day food logs at baseline and week 8.The following questions were used: "How hungry are you right now?/How satisfied are you right now?/ How much could you eat right now?" and were anchored with "Not at all/Nothing" on the left and "Extremely/Vast quantities" on the right. Participants were asked to mark their responses by placing a slash on the 100mm horizontal line.
Visual analog scale measurements were completed before breakfast and before bed for the three days of food intake recording. VAS scores were calculated by measuring millimeters from the left anchor and entered into an Excel spreadsheet for analysis.
The three days were averaged for analysis. All scores were double checked for accuracy.

Body Fat Percentage
Percent body fat was determined using air displacement plethysmography (BodPod model 2000A, Life Measurement Inc, Concord, CA). Standard procedures were used.
Subjects were asked to refrain from physical activity and caffeine consumption for two hours before the test. Subjects were asked to wear a swimsuit or compression suit and a swim cap during the procedure to improve the accuracy of the results. Jewelry, socks and shoes were removed. Body composition measurements occurred during the initial and post-intervention visits to the lab for each subject.

Waist Circumference
Waist circumference was measured in centimeters using a Dritz model 11036 nonstretch, traditional tape measure. The measurement was taken with excess clothing removed. The tape was placed at the level of the umbilicus. Participants were asked to take a deep breath and relax their abdomen as they exhale. Waist circumference was measured to the nearest 0.2 centimeters. Waist circumference was measured fasting, at the pre and post visit for each subject. Measurements were taken in duplicate and averaged, provided both were within ¼ centimeter of each other. Measurements were repeated until two measurements were within given tolerance range.

Height
Height was measured in centimeters using a Seca model 216 stadiometer (Hanover, MD) with excess clothing and accessories removed. The measurement was taken from the floor to the top of the head with feet together and flat on the floor. The trained research assistant ensured the participant's head, shoulders, buttocks, and heels were against the stadiometer and participant's head was in contact with the stadiometer at the Frankfort plane during measurement. The participant was asked to inhale and hold his or her breath. Height was measured to the nearest centimeter. Height was measured during the pre and post visit for each subject. Measurements were taken in duplicate and averaged provided both were within ½ centimeter of each other. Measurements were repeated until two measurements were within given tolerance range.

Weight
Weight was measured in kilograms with an electronic weighing system (Tanita Corporation, Japan). The scale was calibrated using two known 10-kilogram weights before each participant was weighed. Participants were weighed in the center of the scale with shoes and excess clothing removed. Weight was measured to the nearest tenth of a kilogram, fasting, during the pre and post laboratory visit for each subject.
Measurements were taken in duplicate and averaged.

Data Collection
The same researcher performed all measurements to help ensure uniformity in procedures. Each research assistant was trained prior to measurements. Each research assistant was required to show test-retest reliability for trials before data collection began.

Week# Schedule
1  Visit 1: Assessed eligibility, collected baseline data for treatment and control groups, distributed 3 day food records and daily appetite profile  Completed 3 day food record and daily appetite profile  Visit 2: collected additional baseline data for treatment and control groups, randomized to groups, completed fasting VAS, distributed supplement, food records and appetite profile 2-9  8 week supplementation period for treatment groups 10  Visit 3: Collected post-intervention data for treatment and control groups  Completed 3 day food record and appetite profile Data for the treatment and control groups were collected at baseline and at the end of the intervention via questionnaires, anthropometric measurements, food records and appetite profiles. The data provided by the participants was coded to preserve confidentiality.

Statistical Analysis
All numerical data were entered into an Excel spreadsheet (Microsoft Corporation, Redmond, Washington) and transferred into SPSS version 20 (IBM Corporation, Somers, New York) for statistical analysis. All data were double checked for accuracy.
Mean values and standard deviations were calculated. Normality was assessed based on skewness and kurtosis. Descriptive data included gender, age, BMI and body composition at 0 and 8 weeks for each group. Diet records were entered into FoodProcessor SQL (Esha Research, Salem, Oregon) to analyze nutrient composition.
Data from the three diet records were averaged for the three days to obtain a single measurement for diet quality. Dietary quality was analyzed by calculating the AHEI score from the averaged food record. The primary dietary quality outcome was total AHEI score although intake of energy, macronutrients, fiber, fruit and vegetables also were examined. The primary appetite outcomes were lab-assessed fasting hunger and satiety, although free-living daily averages, plus fasting morning and pre-bedtime ratings were also examined. The primary analysis was a three group by two occasion repeated measures analysis of variance in order to observe changes over time within each group as well as a time by group interaction. For appetite data, two groups by two occasion repeated measures analysis of variance was analyzed to observe changes over time with each group as well as a time by group interaction comparing the 2 protein groups to the non-treatment control, since differences between the two protein groups were not found for these data. Pearson correlations of total protein intake with AHEI scores as well as appetite ratings were explored. Effect sizes were reported as partial eta squared values. Statistical significance was set at p<0.05.

RESULTS
Two hundred and four individuals responded to fliers, classroom announcements and mass emails. Of the 204, 78 individuals did not meet the inclusion criteria. Fifty nine individuals had no risk factors for cardiovascular disease. Fourteen individuals had only 1 risk factor for cardiovascular disease. Three individuals were vegetarians who consumed mainly soy protein. One individual was already consuming whey protein on a regular basis. One individual did not meet the age requirements. Of the 126 who qualified, 9 participants met all inclusion criteria and completed all 3 laboratory visits.
Subject characteristics by group for all analyses can be seen in Table 1

Visual Analog Scales
Analysis of variance was used to compare within-group and between-group differences in hunger, satiety and appetite (Table 2). No significant time by group or within group differences were found. Average laboratory measured fasting hunger responses increased for both protein groups, but decreased for the control group (SP=21.2mm, WP=15.0mm, C= -16.2mm). Average laboratory measured fasting satiety responses decreased for both protein groups, but increased slightly for the control group (SP= -21.0mm, WP= -15.5mm, C=2.8mm). Desire to eat increased for both protein groups, but decreased for the control group (SP=3.13mm, WP=20.63, C= -6.7). Average free living fasting hunger responses increased for all 3 groups (SP=5.17mm, WP=12.4mm, C=3.3mm), but was most drastic for the whey protein group. Average free living before bed hunger responses decreased for both protein groups, but increased for the control group (SP= -17.37mm, WP= -6.90mm, C= 17.93mm).

Energy Intake
No significant time by group or within group differences were found for energy intake. Analysis of variance was used to compare within-group and between-group differences in energy intake (Table 3). Average energy intake decreased for all group (SP= -181.8kcal, WP= -822.6 kcal, C= -572.5 kcal), but most drastically for the whey protein group. Although the analysis of variance did not show statistical significance (p= 0.30), the effect size was large (partial eta squared= 0.334) for time by group tests of between subject effects for energy intake (Table 3).

Protein Intake
Following protein supplementation, the experimental groups were consuming significantly more protein than the control group (experimental: 115.3±25.5 grams, control: 73.7±10.6grams, p=.033). There was no significant difference between the two protein groups in terms of protein intake at post measurements. Although the analysis of variance did not show significant differences (p=0.28), the effect size was large (partial eta squared= 0.346) for time by group tests of within group effects for protein intake (Table 3). Pearson correlations were also run to measure the strength of the relationship between protein intake with total AHEI scores and lab-assessed fasting appetite scores (Table 5). There were small correlations between protein intake and lab assessed satiety and desire to eat (Pearson correlations= 0.21, -0.10, p=0.587, 0.799). As predicted, the correlation between protein intake and lab assessed satiety was positive and the correlation between protein intake and lab assessed desire to eat was negative. There were no correlations between protein intake and AHEI scores or between protein intake and lab assessed fasting hunger.

Dietary Quality Scores
Analysis of variance was used to compare within-group and between-group differences in total AHEI scores, fat intake, carbohydrate intake, fiber intake and vegetable intake, as well as percentage of energy intake for protein, fat and carbohydrate (Table 3). No significant time by group or within group differences were found. The average AHEI scores for each group decreased over the intervention period (SP= -1.41, WP= -5.76, C= -1.30), but the average decrease in the whey protein group was much larger when compared to the soy protein or control groups. Fiber intake decreased for all 3 groups (SP= -10.16, WP= -9.02, C= -3.79), but least considerably for the control group. Vegetable intake decreased for the soy protein group by 0.84 servings on average, while intake for the other groups increased very slightly (WP= 0.39, C= 0.22).

DISCUSSION
The purpose of this study was to determine the impact of an 8-week protein supplementation intervention on dietary quality and appetite by comparing groups supplemented with whey or soy protein with each other and with an assessment only control group. Previous protein supplementation studies have made assumptions about dietary changes without actually measuring total dietary intake during interventions.
To our knowledge, this is the first study to examine the effect of whey and soy protein supplementation on dietary quality in the overweight and obese college-aged population. Overall, this study offered exploratory data and generated some insight into how to best implement a protein supplementation intervention in the college-aged population in the future.
Dietary quality and appetite measures were not significantly affected by the addition of a protein supplement for an 8 week intervention. However, there was a large effect size of the intervention, based on partial eta squared for between group effects over time energy intake (.334). Since all three groups decreased energy intake, the partial eta squared value indicates that 33.4% of the changes in energy intake are due to the protein supplementation. Based on the partial eta squared values, over half (57.4%) of the change in AHEI scores were due to the protein supplementation. Since all three groups decreased AHEI, this is a surprising value. For protein intake, percent energy from protein and carbohydrate intake, just under half (47.0%, 42.0%, 45.1%, respectively) of the change in values was due to the protein supplementation, based on the partial eta squared. Future research will need to take this into account when creating interventions to improve diet quality and appetite in overweight and obese college students.
The VAS ratings of hunger, satiety and desire to eat were not consistent across the groups, within the groups or when comparing free living to lab assessed. Flint et al. tested the reproducibility and validity of VAS and found them to be reliable, but this research was measuring single meals and not fasting measures of appetite in the lab and in a free living situation . Contrary to what Mars et al. found earlier in 2012, the VAS measures of appetite did not seem to predict subsequent food consumption . The average energy intake for each of the groups in the current study decreased, while the appetite measure did not reflect this. Similar to the findings of Flint et al, the VAS were not able to predict changes in food intake, especially for the overweight and obese population . The current participants may be "hedonistic" eaters, in which case the VAS measures may not be as accurate since they are used to measure physiological cues rather than responses to pleasure . If individuals do not eat in response to physiological cues, it may be difficult for them to record accurate VAS measures because they may not understand exactly what they should be feeling in terms of hunger or satiety never mind be able to record those feelings. One limitation when comparing the lab assessed to the free living appetite measures, is that during the free living condition, participants woke on their own schedule. If a participant does not typically wake up until noon, and we are measuring his or her fasting hunger levels at seven in the morning, this may have an impact on his or her ability to report accurate appetite levels.
In terms of their dietary quality based on the AHEI scores, the current participants clearly did not follow the dietary instructions. Participants were asked to take the supplement in place of less healthful items, such as sugar sweetened beverages, fast food items, fried foods, et cetera, that provided around 190 kilocalories. Based on this information, the total energy intake should have been similar at pre and post intervention if they simply replaced the 190 kilocalorie item with the 190 kilocalorie protein supplement. The energy intake decreased for all three groups. The participants decreased their fruit and vegetable intake, which in turn decreased their fiber intake. Instead of decreasing less healthful items, the participants replaced the already limited amounts of fruits and vegetables they were consuming with the protein supplement. One limitation of the AHEI is the fact that it does not account for total kilocalorie intake. Individuals who consumed more calories had higher AHEI scores just because they were consuming larger amounts of food, so there were more opportunities for that individual to consume food items that would increase their total dietary score. The AHEI does not take points away to less healthful choices, such as fried foods and sugar sweetened beverages. If two individuals had all the same food items, but one consumed just water and the other consumed just sugarsweetened beverages, they would have the same total score because sugar intake is not included in this diet quality index. This is clearly a limitation of the total AHEI scores.
In terms of dietary quality, the current participants confirmed the findings of Racette et al., who found that college students do not meet the guidelines for dietary patterns . For fruit and vegetable intake, the average participant was consuming 0.5 and 1.59 serving respectively. These numbers clearly do not meet the recommendations set forth by the Centers for Disease Control. On average, the female participants should be consuming around 2 cups of fruit and 2.5 cups of vegetables each day. On average, the male participants should be consuming around 2 cups of fruit and 3. While the amount of protein did increase for all individuals in the experimental groups, it did not have a significant impact. There may not have been a great enough increase in protein intake. Weigle et al. found that adults decreased their energy intake (-441±63kcal/day) when protein was increased from 15% to 30% of total energy intake . The current participants in the experimental groups were consuming 19±0.04% of energy from protein at the end of the intervention period.
Paddon-Jones et al. found that metabolic profile is improved once protein intake reached greater than 25% of total energy intake . These studies did not examine the effects of a protein supplement, but rather used food items to increase the subject's protein intake.
There are some limitations that are worth mentioning. This study's greatest limitation was the small group sizes. Recruitment of more subjects and a more diverse population are needed for future studies. Significance for some of the results may have been achieved with a larger sample size. Also, subjects were all overweight or obese individuals. Previous research suggests that overweight and obese individuals are most likely to underreport food consumption, especially in the free-living environment.
Some of the reasons for underreporting include trying to please the researcher, forgetting, undereating on test days or not recording what they actually ate (de Castro, 2004). This may have occurred in our study resulting in lower recorded energy intakes. However, all subjects reported energy intakes above the cutoff criteria established by  in previous studies suggesting that the reported energy intakes of these participants are plausible values. Another limitation is the use of food logs. Subjects were instructed to record the three days prior to their laboratory visit. This resulted in the majority of subjects recording 3 weekdays. Typically 3 day food logs include 2 weekdays and 1 weekend day to create a better overall picture of the participant's normal eating habits. Future studies may want to use more accurate methods for measuring energy intake. Multiple pass 24-hour recalls with the Nutrition Data System for Research, may be more accurate and reliable than 3 day diet records (Conway, Ingwersen & Moshfegh, 2004), but that has not been confirmed in the overweight or obese college aged population.
There are also some strengths of this study. All subjects were tested alone to minimize outside influences. This also allowed for personalized dietary information to be given to each participant. The three groups contained the same number of people as well as the same percentage of males and females. Also, a control group was used to compare both within and between group data. The same researchers conducted all of the participant visits to ensure uniformity in measurement techniques. Data were collected during both the fall and spring semesters.
If this study were to be repeated, there are a few changes that should be made.
Clearly, more subjects would need to be recruited to increase the likelihood of finding statistical significance. Participant visits should be based around the participant's typical schedule, instead of bringing all subjects in first thing in the morning. More accurate measures of energy intake should be used. Overall, this research gathered valuable observations of this intervention in the overweight and obese college-aged population. A protein supplement may not be able to improve the diet quality or appetite levels of overweight or obese individuals, even with dietary counseling, but this is important information to bring to future research studies. Ideally with a larger sample size, detailed insight into the effectiveness of a protein supplementation intervention can be assessed.       Zhang, X., Shu, X. O., Gao, Y. T., Yang, G., Li, Q., Li, H., . Soy food consumption is associated with lower risk of coronary heart disease in Chinese women. Journal of Nutrition, 133, 2874-8.

Introduction
The United States has seen a substantial increase in obesity, cardiovascular disease, and diabetes partially related to an unhealthy diet. Both dietary quality and satiety are important factors that can help modify obesity and associated health conditions, but the impact of various interventions on dietary quality and satiety has not been clarified. Protein supplementation has been shown to improve satiety in overweight and obese individuals, but research regarding improvement in diet quality is lacking. Further, little work has explored the relative influence that different types of proteins may have on such outcomes.

Obesity
Obesity has been on the rise in the United States since the late 1980s . In 2008, over 74% of the adult population in the United States was overweight, obese or extremely obese   .
Obesity is a complicated disorder associated with many chronic diseases such as coronary artery disease, type II diabetes, hypertension, dyslipidemia, stroke, sleep apnea, and some cancers . Obesity is also extremely expensive. According to the Centers for Disease Control, obesity costs $147 billion in obesity-related medical care annually . Also, overweight and obese individuals may find themselves with lower health-related quality of life, compared to their normal weight counterparts . Obesity is partially due to chronic positive energy balance , related to a diet lacking in balance, variety and moderation (U.S. Department of Health and Human Services, 2010). Since obesity has such a negative impact on individuals and society as a whole, it is important for researchers to explore approaches to improve this current epidemic.

Health Risks of Obesity
Obesity is related to many health problems including as coronary artery disease, type II diabetes, hypertension, dyslipidemia, stroke, sleep apnea, depression and some cancers .
Previous research has shown that overweight individuals live shorter lives  Obesity also leads to greater risk of coronary artery disease, as well as elevated total cholesterol and hypertension . While looking at the relationship between BMI before age 30 and coronary heart disease, Owen et al.
found that a one kgm -2 rise in BMI was positively associated with an 8% increase in the risk for coronary heart disease .
Obesity also impacts quality of life due to increased incidence of body pain, fatigue and physical limitations . Obesity is associated with higher rates of depression and lower self esteem . Also, reproductive function is reduced with obesity .

Obesity in College-Aged Individuals
According to the National College Health Risk Behavior survey, 35% of college students are overweight or obese, which may be due to lack of physical activity and unhealthy diets . College students are within the 18-29 year old population, which is where the greatest increase in weight gain and obesity has been observed . The college environment can be conducive to overconsumption of energy dense food . Levitsky et al. measured weight changes from high school to the first few months of college and reported that between 58% and 71% of the total variance of weight gain was due to increased food consumption and unhealthy eating behaviors . Levitsky et al. identified the following as factors contributing to unhealthy weight gains: 'all you can eat' facilities, high fat food consumption, evening snacks, and increased consumption of energy dense foods with low nutrient density .
Similarly, data from the CARDIA study, a prospective, epidemiologic investigation of cardiovascular risk in 5,115 young adults between the ages of 18 -30 years, found that all groups experienced significant increase in prevalence for all overweight categories (BMI >25kg/m 2 ) and also decreases in the normal weight category from baseline to year 10 ( . These subjects were studied from 1985-1996. These young adults gained an average of .06-1.19kg/year over the ten years. This study shows that weight gain is much more common than weight stability or weight loss and the largest weight gains occur more often among individuals who were overweight at baseline compared with those of normal weight .
Another study found obesity rates increased from 14.7% to 17.

Appetite
Appetite is what drives an individual to find, choose and consume food . There are numerous ways to measure appetite, including subjective ratings, food intake, or physiological markers . Under normal circumstances, satiety occurs following food ingestion to inhibit further consumption. Between eating episodes, satiety typically prevents food intake and delays the start of the next meal or snack . Satiety can be affected by many factors, including energy density, weight, volume, macronutrient composition, appearance, satisfaction and palatability of certain foods or meals . In terms of public health and controlling obesity, satiety is an important factor to consider .
Individuals consume food for a number of reasons. From a physiological standpoint, the body needs food and these needs must be satisfied for su rvival.
When energy availability is low, the body will respond by increasing motivation to eat or appetite . One problem with homeostatic controls of appetite is that they may be overtaken by hedonic drivers of consumption.
Individuals who are considered "hedonistic" eaters may consume food in response to pleasure rather than the physiological cues the VAS are intended to measure . Some nonphysiological factors that influence energy intake are social circumstance, dietary restraint, availability, properties of the food, and subjective judgment of the food .

Measuring Appetite
VAS are one method used to measure appetite. Previous research has shown that changes in appetite can be detected by VAS   . Other research has shown that changing eating behavior is much more complicated than just improving subjective appetite scores . Visual analog scales do not always predict changes in food intake, especially for the overweight and obese population .
The debate is still out on whether or not VAS can predict energy intake.

Diet Quality
More recent diet and health related epidemiologic studies have switched their focus from single nutrients to overall indicators of diet quality and patterns . At this point, there is no universally accepted measurement of dietary quality, nor is there a universally accepted definition. According to Fung et al, the purpose of diet quality indices are to measure and guide people toward a dietary intake that will promote health and prevent disease . Dietary quality has been measured using various indices including the Healthy Eating Index (HEI), Diet Quality Index,

Recommended Food Score, Dietary Variety Score and the Alternate Healthy
Eating Index (AHEI) . Researchers will need to continue to look for the ideal combination of dietary factors and also the best way to assess whether the population is adhering to this ideal combination.

Diet Quality & Chronic Disease
Poor diet quality, as defined by the Healthy Eating Index, is associated with obesity and overweight status. Guo et al. found that average HEI scores computed from 24 hour food recalls were significantly lower for obese individuals compared to their normal weight counterparts. On average, obese individuals scored 62.3, while normal weight individuals scored 63.6. This was found to be statistically significant (p=.04) This study was a cross sectional analysis of 10,930 adults who completed the Third National Health and Nutrition Examination Survey . Improving diet quality could influence weight status in some individuals . High dietary quality as measured by the AHEI, has been correlated with lower risks for major chronic diseases . Akbaraly et al. found that individuals in the top tertile for AHEI score had about a 25% lower chance of all-cause mortality and 40% lower chance of mortality from cardiovascular disease, compared with individuals who scored in the bottom tertile for AHEI score . This study included semiquantitative food frequency questionnaires for 7319 participants as well as an 18 year follow up.
McCullough and Willett collected dietary intake data from two large cohorts of men and women between 1984-1990 and found that participants whose dietary intakes were closest to meeting the AHEI goals had a 20% and 11% lower risk of major chronic disease. More specifically, these individuals in the top quintile for AHEI score, had significantly lower risk for cardiovascular (39% in men, 28% in women), compared to the lowest quintile  They found that all of the diet quality scores were significantly correlated to each other. Also, once researchers adjusted for age, body mass index, alcohol intake, physical activity, smoking status, and energy intake, only the AHEI and aMed scores were associated with significantly lower concentrations of biomarkers of inflammation and endothelial dysfunction. Both endothelial dysfunction and inflammation are related to diseases such as atherosclerosis and diabetes. The researchers believed the AHEI and aMED scores were associated with lower concentrations of inflammatory and endothelial dysfunction biomarkers because of their focus on high consumption of fruits and vegetables, whole grains, nuts, and fish and moderate alcohol .

Alternate Healthy Eating Index
The Alternate Healthy Eating Index was created to improve the Healthy Eating Index, by targeting food choices and macronutrient sources that have been shown to decrease chronic disease risk . The AHEI focuses on vegetables, fruits, nuts & soy, ratio of white to red meat, fiber, trans fat, ratio of polyunsaturated to saturated fat, multivitamin use and alcohol consumption.
Each of the components of the AHEI were included because of their ability to decrease risk for cardiovascular disease. Adequate vegetable intake has been linked to decreases in chronic disease risk . All vegetables except for potatoes are included in the Alternate Healthy Eating Index. Five servings of vegetables per day were found to be ideal based on the current dietary guidelines . Fruit intake has been linked to decreases in cardiovascular disease . Four servings of fruit per day were found to be ideal based on the current dietary guidelines . Nuts and soy protein have both been associated with lower risk of cardiovascular disease .
McCullough et al. found one serving of nuts and soy protein per day to be ideal . The white to red meat ratio was included based on the knowledge that fish and poultry have been linked to lower risks of coronary heart disease and cancer, while red meat and processed meats have been shown to increase these risks .
Fiber has been associated with decreased risks of coronary heart disease and also stroke . Trans fatty acids have been shown to raise LDL cholesterol, lower HDL cholesterol, and increase coronary heart disease risk . A high polyunsaturated fat intake, compared to saturated fat intake has been shown to lower coronary heart disease . Long-term folate intake and supplementation have been associated with decreased coronary heart disease and cancer. Folate is typically found in multivitamins. . Moderate alcohol consumption has also been associated with a lower risk for cardiovascular disease. McCullough et al. defined moderate intake as 1.5-2.5 drinks per day for men and 0.5-1.5 drinks per day for women .

Diet Quality of College Students
Previous research has shown that college students as a whole, do not comply with recommendations for healthy diet practices ). Another study observed a sample of students at a large university to evaluate their nutrition knowledge, beliefs and practices. The study found that most of the students had a good understanding of basic nutrition, but the majority (69%) of them consumed less than one serving of fruit per day and about half (43%) of the students consumed less than one serving of vegetables per day ). More research is necessary to determine what causes these habits. One researcher identified skipping breakfast, snacking on chips or sweets, consuming sweetened beverages and consuming fast food when short on time, as behaviors related to weight management in college students that need to be addressed . College students are introduced to a large variety of energy dense, less healthful foods, and new dietary patterns when they transition from high school to college. It is important for researchers and healthcare professionals to determine the causes of these unhealthful behavior changes and find ways to improve them.

Improving Diet Quality
Previous research has shown that dietary patterns following the Mediterranean-style diet result in higher dietary quality. These dietary patterns include more whole grains, vegetables and fruits. The Mediterranean diet also typically includes higher levels of fat intake from plant sources. The current Western-style dietary pattern, which is composed of more red meat, processed meat, sugar sweetened beverages, sweets, refined carbohydrate and potatoes leads to lower quality diets and obesity   . Future work should focus on improving the diet quality of overweight and obese college aged individuals through nutrition education and restructuring current dietary patterns.

Benefits of protein intake
Protein intake greater than the recommended 10-15% of total energy intake maybe a strategy for successful weight loss and also the prevention of weight gain following weight loss . Some of the additional benefits, beyond improved satiety, include increased thermogenesis and maintenance or accretion of fat-free mass .
Improved thermogenesis may improve energy expenditure. When protein intake is greater than 25% of energy intake, some individuals retain lean muscle mass and improve metabolic profile possibly due to improved muscle protein anabolism .

Protein Intake & Satiety
Individuals who find difficulty in controlling their appetite may find satiety-enhancing foods beneficial to help decrease the urge for consumption . Protein is known to be the most satiating macronutrient , related to increased diet induced thermogenesis ). This makes protein intake an important factor in the context of weight loss and management . Higher protein diets have been shown to decrease appetite, as protein increases satiety and metabolic rate . Previous research has shown that protein-induced satiety from high protein ad libitum meals can last from 1 to 6 days, up to 6 months .
In a study conducted by Dougkas et al, a morning snack with an average of 12 grams of protein decreased appetite and subsequent lunch intake compared to the control group . In a similar study, additional benefits were seen when the protein snacks contained 24 grams of protein . Researchers concluded that a small, high protein snack consumed most days may delay or prevent snacking or overeating for the latter part of the day .
Both Douglas and Dougkas, used yogurt snacks and recruited only healthy women between the ages of 18 and 50.
In a study conducted by Poppitt et al, researchers discovered that having overweight and obese women consume a protein-enriched water beverage containing 5 to 20 grams of protein prior to an ad libitum buffet lunch, would result in improved feelings of fullness and less hunger for the six hours following the preload compared to a water control condition. Forty-six women participated in the double-blind cross over study for each of the four beverage conditions, including a water control, 1%, 2% and 4% protein by weight beverage conditions. Researchers did not see a significant change in energy intake .
On the other hand, Wiegle et al, found that by increasing protein from 15% to 30% of total energy intake in 19 healthy adults, average energy intake decreased (-441±63kcal/day) while satiety levels were maintained during a 12 week intervention, compared to baseline food logs and visual analog scales . Most studies measuring the effects of protein supplementation on appetite focus on healthy individuals, not the overweight and obese population. Many of these studies look at the postprandial effects of the supplementation following a laboratory meal and not during a free-living situation . Future research should focus on the ability of protein to improve satiety in a free living environment, rather than just in the laboratory. Also, further work should examine effects in overweight and obesity population.

Satiety & Protein Source
In addition, different protein sources may affect the body differently in regards to satiety . In a study conducted by Alfenas et al. (2008), researchers found that soy protein increased thermogenesis, whey protein decreased respiratory quotient and casein decreased energy intake. This study only included normal weight subject s for the four 7-day experimental, crossover sessions. One strength of this study was that the amount of protein added to the diet was based on the individual's weight in kilograms. The researchers added 2 grams of protein per kilogram of bodyweight to the test meal of milkshakes with crackers, cookies or cake . Both whey and soy proteins have been shown to aid in weight loss, lower blood pressure, improve lipid profile, and reduce overall risk of cardiovascular disease in some populations , but have not been systematically compared for their effects on satiety in overweight and obese young adults.
In contrast to the above studies, Lang et al. did not observe any differences in satiety when comparing protein source, which included egg albumin, casein, gelatin, soy protein, pea protein and wheat gluten. Participants included 12 healthy subjects who each consumed the 6 protein-manipulated lunches . This may be due to differences in methodology. Lang et al. added the protein directly into the lunch meal, while other studies incorporated protein as a preload to an ad libitum lunch meal. Also, the researchers noted that fiber was not controlled during some of the previous studies, which could mask the satiating capacity of the protein intake )

Protein Supplementation
Some protein supplementation research has been completed in the college population , but the majority of the work looks at the effect on physical performance in healthy subjects and examined few nutrition parameters (Fluegel, Shultz, Powers et al., 2010).
Researchers have assumed that the diets of their subjects either stay the same or improve when they add a protein supplement, which may not be accurate. Neither soy nor whey protein supplementation has not been addressed in the college population in regards to improving dietary quality or satiety.

Obesity & Reporting Energy Intake
Underreporting of energy intake is a reoccurring challenge in nutrition, especially when self-report assessment methods are utilized. Underreporting occurs when people report estimated food intakes that are lower than their true energy intake. Poslusna et al. define underreporting as the discrepancy between reported energy intake and measured energy expenditure without any change in body mass during the observation or reference period . Researchers have suggested that reported energy intake can be used to assess an individual's reported energy intake compared to their energy requirement .
According to previous research, overweight and obese individuals are more likely to underreport their energy intake than their normal weight counterparts. Pietiläinen et al found that obese individuals significantly (p=0.036) underreported their energy intake by an average of 764 kilocalories . Another study, conducted by Buhl et al, found similar results. This study used doubly labeled water and reported that each of their ten overweight participants underreported energy intake. This was used to explain the participants inability to lose weight after being placed on energy restrictive diets and not reporting any weight loss. All participants had previously reported consuming less than 1200 kilocalories per day .
Another study found that obese individuals underreport their energy intake by 20-50% . The higher the BMI of the participant, the greater the risk of underreporting of energy intake at any given meal ). Researchers have not been able to determine how to prevent this underreporting. Clearly, accurate methods of determining free living energy intake for overweight and obese individuals still needs to be discovered.

Conclusions
Overweight and obesity are preventable causes of morbidity and mortality that affect a vast majority of the United States adult population