Effects of Nutrition Education Supported by Mobile Application on Weight Loss and Quality of Life: A Randomized Controlled Trial

Background Obesity is a growing health problem which affects people from all age groups all over the world. Obese people do not feel motivated enough to change their lifestyle behaviors. Mobile applications can be used to motivate people. Objective To evaluate effect of nutrition education supported by MOtiVE mobile application on weight loss and quality of life (QoL). Methods In this randomized-controlled study, 79 overweight/obese adult patients who presented to University between 2018 to consult were included. All the participants were provided a weight-loss diet program by the dietitian. Then, participants were randomized to experimental and control groups. During the rst interview, all participants completed the questionnaire and anthropometric measurements were done. BMI, the scores obtained from different Quality of Life scales and Healthy Eating Index (HEI) were the dependent variables. The participants in the experimental group were provided with daily messages for 3 months via MOtiVE mobile application designed solely for this study. All the participants were asked to present three months later for a follow-up appointment. Using SPSS 25.0, change in BMI, QoL scores and other variables within both groups was assessed via Wilcoxon signed-rank test and McNemar chi-square test. p < 0.05 was considered statistically signicant.


Background
Obesity is a multifactorial and chronic disease characterized by excessive accumulation of fat in the body resulting from the interaction between genetic and environmental factors. According to World Health Organisation in 2016, more than 1.9 billion adults were overweight and over 650 million were obese (1). In weight loss programs, the most important factor is lifestyle changes. Internet-based programs are potential supporters for individuals participating in weight loss programs because the Internet is easy to access. Studies conducted on this issue have shown that such programs, although short-term, lead to positive changes in diet, physical activity level, glycemic control, anthropometric measurements and biochemical parameters (2)(3)(4)(5). In several studies, low self-esteem and poor body image were determined to be associated with obesity (6,7). Providing motivational support to achieve lifestyle changes can play a signi cant role in maintaining weight loss by increasing self-esteem. The main purpose of the present study was to compare the effect of standard education on weight loss, quality of life (QoL) and healthy eating behavior with that of nutrition education supported by mobile application. It was assumed that the mobile applications would increase people's motivation for and adaptation to diet.

Trial Design
The present study were conducted with patients who presented to the _____ University Hospital Endocrine&Metabolic Diseases Outpatient Clinic between March-September 2018 to consult a dietitian and to receive routine nutrition therapy, aged between 18-64 years and Body Mass Index (BMI)≥25. The participants were rst given the standard nutrition training by the dietitian in the outpatient clinic, and then person-speci c diet program. After this, the patients who gave their consent to participate in the study were randomly assigned to the case and control groups. Then, anthropometric measurements of all the participants were performed and the data collection forms were lled in. The "MOtiVE" mobile application for smart phones designed speci cally for this study was installed on the smart phones of the cases, and text, visual or video messages were sent to them via this application for three months. The participants in the control group underwent the routine procedure. In line with the routine practice of the outpatient clinic, all the participants were told to present three months later for control appointments. At the end of December 2018, at the end of the three-month follow-up period, the data collection process ended. In a systematic review conducted by Aguilar-Martınez et al, showed that most of the trials last for 2-4 months(8).
Participants and Recruitment Of the patients, those who did not have a smart phone with internet connection, underwent bariatric surgery, took medication or practice a special diet for thyroid problems, diabetes, celiac, gout and kidney diseases and whose body fat analysis was not ful lled due to the presence of a pacemaker, prosthesis etc. were not included in the study.
According to the results of a study conducted by Allen et al, 25 participants in each group was enough to reach 80% power (9). In a meta-analysis conducted by Mateo, revealed that many studies had 40-45 participants in each group, some of them being less(4). When we consider 80% power with a %5 error and medium effect size (d:0.5) we had to have 51 participants in each group calculated via G-power. When we evaluate admissions to the outpatient clinic, we saw that each week we can have 4 or 5 new participants. Thus, considering this data, we decided to take 40 participants in each group. The owchart of the study is given in Figure 1.
In the present study, randomization was carried out in the following block allocation sequence: AABB, was measured with a non-elastic standard measuring tape, with the individual wearing light clothes, standing still, in an upright position and with arms open sideward. WC was measured from the midpoint between the distal border of the lowest rib and the superior border of the iliac crest (10). Neck circumference was measured by placing the inelastic measuring tape around the neck from the point just below the laryngeal prominence.
QLSSF-36: The scale consists of social functioning, physical functioning, role limitations due to physical problems, mental health, role limitations due to emotional problems, energy/vitality, and perception of general health and pain subscales. The higher the score is the better the quality of life is (11,12).
WEL Test: It was consisted of 20 questions about negative feelings, food accesibility, social pressure, physical disturbance and positive activity. The higher the score is the better the weight e cacy lifestyle is (13,14).
HEI: It was developed by the United States Department of Agriculture to investigate the quality of diet in Americans (15). The new HEI-2015, which was published in 2018, consists of 13 components related to diet. The increase in scores indicates a positive development (16). Nutritional values for individuals based on their 24-hour food consumption records were calculated using the _________ (______ dietary data system) and measurements were calculated based on Dietary Guidelines for _______, 2015 (17).
RSES: The rst 10 items of the scale are used to measure self-esteem. Self-esteem increases as the score obtained from the RSES decreases (18,19).
OWLQOL: The higher the score obtained from the scale is, the higher the quality of life of the person is (22,23).
All of the scales were validated in ___________and found to be reliable to use in ________.
MOtiVE Application for Mobile Phones: The program was developed by the researchers with taking technical support from a software engineer, speci cally for this study and it is compatible with both Android and IOS. The messages were sent to the participants in the case group once a day for three months. Messages sent to the phone screen every morning at 9:00 AM were either text, visual or video messages. The visual messages designed to draw the attention of the reader included both a text and an illustration ( Figure 2).
By sending text messages like "Sunday is the only day when most of us do not work. You will be more comfortable during the weekdays if you do your shopping and prepare meals on the weekend. Prepare your meatballs this Sunday and freeze them….", practical solutions were offered to the participants. By sending text messages like "Lifestyle changes improve your quality of life and enable you to lead a better quality of life. Improve your quality of life with diet and regular exercise. Thus, do not give up..." the participants were motivated.
Video messages were uploaded to YouTube and a link address (e.g.What is an antioxidant? https://www.youtube.com/watch?v=TIaQq1CbNBM) was sent to the cases, which enabled them to access. The videos were also subtitled so that the watcher could get the message in noisy environments.
The registration screen of the MOtiVE, home screen layout, samples of messages sent for 90 days are given in Additional les.
One of the features of the MOtiVE is to calculate for how many days the participant was following the program and to send the message speci c to that day to the person's phone. Thanks to this feature of the MOtiVE, it was possible to send a message like "Congratulations, you have been on a diet for a full month…" to the person who participated in the program for a month. So the program was made personspeci c. Another feature of the MOtiVE is that only one-way communication was possible in order to avoid bias. The application enabled only the researchers to send messages to the participants but did not allow the participants to communicate with the researchers. In this way, standardization was established between the participants in terms of determining their knowledge and interest. Another feature of the MOtiVE is to report whether the participants see the message of that day, which enabled the researchers to nd out whether the participant has read the messages.

Statistical Analysis
In the analysis of the study data, IBM SPSS Statistics (version 25; IBM, New York, NY, USA) was used. Continuous variables were presented with mean±standard deviation. At the baseline, the homogenously randomization of the participants to the case and control groups was investigated using the independent samples t-test (Mann Whitney-U test if parametric condition could not be met) and chi-square tests. The e cacy of the intervention was assessed using Wilcoxon signed rank test separately in the case and control groups. p<0.05 considered signi cant.

Ethical issues
Study was conducted in accordance with the Declaration of Helsinki and ethical approval was obtained from ______ University Medical Research Ethics Committee (no:17-7.1/14, 08.08.2017) and the written consent was obtained from the patients. The study was recorded in clinical trials (no: ClinicalTrials.gov NCT04026971).

Results
The number of the patients who took the rst test was 79 (cases:39, controls:40). The number of individuals who took the nal test was 38 (cases:20, controls:18). According to the feedback provided by MOtiVE, all participants read all the messages.
Of the participants, 90% in the case group and 83.4% in the control group were women. There was no signi cant age difference between cases and controls (34.7±14.0 years vs. 38.7±13.5 years; p=0.349). The analysis performed to test whether the participants who took the nal test had been homogenously randomized to the case and control groups at the baseline revealed that there were no signi cant differences between the groups in terms of their sociodemographic characteristics, anthropometric measurements, or the scores they obtained from the RSES, OWLQOL, HLSB-II scale, WEL test and HEI and their sub-dimensions except for the nutrition subscale of the HLSB-II scale. The participants in the control group obtained signi cantly higher scores from the nutrition subscale of the HLSB-II scale (Case group: 20.9±3.6, Control group: 23.6±4.2, p=0.043). Therefore, it can be said that the participants were homogenously distributed to the case and control groups.
Anthropometric Measurements: Although weight, BMI and neck circumference measurements of the participants signi cantly decreased in both groups at the end of the 3 months, these differences were more dominant in the cases. Waist circumference decreased signi cantly only in the cases. The changes in the anthropometric measurements of the participants at the end of the three-month follow-up were given in Table 1. The comparison of all the data collection tools used in the present study is presented in Table 2.

Discussion
The knowledge score, diet quality score, and success score indicating success in maintaining a healthy life increased signi cantly in the cases. There was a signi cant correlation between the frequency of using the application and high success score (p<0.01) (24). Similarly in their randomized controlled study aimed at weight loss, Kevin Patrick et al. sent SMSs and/or multimedia messages to the participants in the case group 2-5 times a day for 16 weeks. At the end of the 16 th week the weight loss in the cases was signi cantly higher (p=0.02). It was concluded that SMSs and multimedia messages could promote behaviors supporting weight loss in obese adults (25). In present study, all participants took nutritional trainig rst with their personal diets and then randomized to cases and controls. In a meta-analysis including 14 randomized controlled trials to investigate whether internet-based interventions were effective in empowering patients concluded that these interventions yield to positive improvements. On the other hand, in 3 studies using general self-e cacy scales and in 1 study using the RSES conducted to assess self-esteem, no changes were observed. The comparison of face-to-face interviews and internet-based interventions demonstrated that no signi cant differences were observed in self-esteem (26). In the present study, cases received messages addition to the routine nutritional therapy.
Unlike the meta-analysis, a signi cant decrease in RSES scores thus, a signi cant increase in self-esteem was observed in cases in the present study. However, according to the analysis of the self-e cacy scores obtained from the WEL test, we also could not show signi cant changes in both groups. It is thought that internet-based interventions can be used to improve self-esteem, lifestyle behaviors and quality of life of patients but not self-e cacy.
Although the use of a mobile application led to changes in health behaviors, the mechanisms by which these applications facilitate behavior change are generally not known. Joshua H West et al. conducted a cross-sectional study including 217 participants. The participants gave their feedback about their diet and nutrition applications in the last 6 months and most of the participants agreed that the application increased their dietary motivation, improved their self-e cacy, and increased their willingness to set dietary goals and to reach the target. Therefore, it was concluded that diet and nutrition-related practices focusing on the improvement of motivation, willingness, self-e cacy, attitude, knowledge and goal setting might be particularly useful (27). Jacobs et al demonstrated the importance of adherence to intervention on weight loss in a large sample using a smartphone application Noom tracking individual self-monitoring and showed that after three months, signi cant reduction in BMI was accomplished. They concluded that smart phone application use can induce weight loss associated with adherence (28). The present study was also aimed at improving healthy lifestyle behaviors via inducing motivation. At the end of the present study, there was a signi cant improvement in the mean scores BMI as well as from the overall HLSB-II scale and only its nutrition and stress management subscales.

Limitations and Strengths
The mobile application MOtiVE and its unique features solely designed for this study can be considered as the main strenght to this study. Personel feedback about messages, personel timing and sending messages according to enrollment time makes this programe tailored to participant. Moreover, one-way communication ensured standardization about knowledge and motivation. The main limitation of this study was losses during follow-up period. Nearly half of the participants didn't complete the study. In cases, even though they fallow the messages, some of them refused to come for post-test and nal measurements. This may be due to conducting this study in a university hospital where control appoinments could only be given for three months later. This showed us that people need stronger motivations to continue follow-ups.

Conclusions
In conclusion, it is thought that provision of mobile application based nutrition education to overweight and obese individuals in addition to the routine nutritional therapy may lead to improvements in anthropometric measurements, self-esteem, healthy lifestyle behaviors, quality of life and healthy eating habits of the participants, and it might help to achieve the targeted weight loss. As the number of dietary and nutritional practices continues to increase, to ensure healthy lifestyle behavior changes, application developers together with health professional should consider integrating the appropriate theoretical structures into the newly developed mobile applications Because these types of mobile applications are easily accessible, such applications should be created and made available to the public free of charge in the protection of public health, and prevention and reduction of obesity. All the authors declare that they approved the nal version to be published.

Ethical issues
Study was conducted in accordance with the Declaration of Helsinki and ethical approval was obtained from ______ University Medical Research Ethics Committee (no:17-7.1/14, 08.08.2017) and the written consent was obtained from the patients. The study was recorded in clinical trials (no: ClinicalTrials.gov NCT04026971).

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Additional les.docx