The Integrated Nutrition Education on eHealth Intervention and Its Effectiveness on Improvement of Anthropometric Status and Behavioural Outcomes of Obese Adults: A Systematic Review Intervention eHealth Terintegrasi Pendidikan Gizi and Efektivitasnya terhadap Perbaikan Status Antropometri and Outcome Perilaku Dewasa Gemuk : Tinjauan Sistematis

ABSTRAK Latar Belakang: Masalah gizi lebih telah menjadi isu epidemik tingkat global. Pendidikan gizi, sebagai salah satu cara yang efektif diterapkan dalam Intervention pencegahan dan penanggulangan obesitas semakin berkembang dengan memanfaatkan media eHealth sebagai saluran penyampaian pesan. Tujuan: Mengidentifikasi studi yang mengintegrasikan pendidikan gizi dengan teknologi eHealth dalam pencegahan dan penanggulangan obesitas serta bentuk teori perilaku dan teknik perubahan perilaku (Behaviour Change Technique/ BCT) serta efektivitasnya terhadap perbaikan status antropometri and outcome perilaku pada usia dewasa gemuk. Ulasan: Pencarian artikel dilakukan menggunakan data base online PubMed, Science Direct, ProQuest, and pencarian manual menggunakan Google Scholar pada jurnal bereputasi sepuluh years terakhir dan diperoleh 17 artikel berbahasa Inggris. Terdapat dua bentuk saluran komunikasi dan aktivitas yang digunakan untuk penyampaian pesan pendidikan gizi, yaitu saluran interpersonal dan media massa berbasis internet dengan pendampingan dan atau tanpa pendampingan. Aplikasi seluler merupakan media eHealth yang paling banyak digunakan sebagai alat penyampaian pesan. Teori yang paling sering The Integrated Nutrition Education on eHealth Intervention and Its Effectiveness on Improvement of Anthropometric Status and Behavioural Outcomes of Obese Adults: A Systematic Review


INTRODUCTION
Obesity is a nutritional status condition with a Body Mass Index (BMI) score of >25 kg / m2 which is categorized into overweight (more weight) and obesity 1 . The World Health Organization (WHO) suggests that obesity has become an epidemic issue at the global level as it has continued to triple in prevalence from 1975 to 2016. The trend of increasing the nutritional status of fat in adult individuals aged >18 years also occurred in Indonesia. The prevalence of overweight has almost doubled from 2007 (8.6%) to 2018 (13.6%) and cases of obesity also doubled from 2007 (10.5%) to 2018 (21.8%) 2 .
The results of studies from several countries mentioned an increase in BMI scores in individuals during the covid-19 pandemic. On the other hand, the severity of covid-19 was further exacerbated by the presence of obesity 3,4 , especially in the younger age group 5 , so currently, the implementation of intervention programs related to the prevention and prevention of obesity during the covid-19 pandemic needs to be a concern. In addition to BMI, measurement of the anthropometric status of waist circumference and body composition (example: decreased fat mass) should be considered in helping evaluate the effectiveness of nutritional intervention programs. This is because BMI cannot describe body composition. After all, it is not able to distinguish between muscle mass and fat mass and cannot distinguish the distribution of fat in the body which is an important factor of obesity-related diseases [6][7][8] .
The main factor of obesity is mainly an unhealthy diet and lifestyle. If these two lifestyles are not treated immediately will cause an increased risk of noncommunicable diseases (NCD) including type 2 diabetes mellitus, cardiovascular disease, some cancer events, and even premature death 9 . Lifestyle change interventions through improved eating behavior and physical activity are effective approaches to addressing the incidence of obesity and improving health outcomes 10,11 . It takes an approach to be able to increase individual knowledge and awareness of the risk factors of disease events accompanied by ways to realize behavioral changes by the desired goal through nutrition education. The scope of the subject matter in nutrition education is very broad, covering issues around food, food safety, nutrition, and physical activity that encourage a person to live more actively with a healthy diet 12 . In several studies, nutrition education interventions in adults with obesity improve eating habits as well as blood serum markers and body anthropometry 13 .
Nowadays, nutrition education media is currently growing, ranging from face-to-face methods to the use of eHealth as a message delivery channel. EHealth is defined as health services and information delivered through the internet and information technology that make a health intervention more accessible. The current condition of the covid-19 pandemic forces every health program planner, especially in the field of nutrition education to adapt to technology to minimize barriers related to distance 15 . Therefore, this systematic review study aimed to answer the following research questions: (1) what were the characteristics of studies that integrate nutrition education in eHealth media? (2) What were the nutritional education materials, use of behavioral change theories, and techniques used in the study? (3) Could integrate nutrition education in eHealth-based intervention media provide the improved anthropometric status (BMI, waist circumference, and percent body fat), changes in eating behavior, and physical activity?

METHODS
This systematic review used three electronic databases: PubMed/Medline, Science Direct, and ProQuest. Keywords used in article searches were a combination of the words nutrition education, eHealth, adult, obesity, body mass index, body composition, and weight loss by using boolean operators "AND" and "OR" to provide narrower and more productive searches conducted in the period April 1 st -June 15 th , 2021. Manual searches for relevant studies were also performed from the article reference list in Google Scholar. The selected article was an English-language article in the last ten years (2011)(2012)(2013)(2014)(2015)(2016)(2017)(2018)(2019)(2020)(2021) and was fully accessible and published from a reputable journal indexed Scopus based on Scimago journal rank (SJR). The reason for the selection of the last ten years was to consider that the results of this library review were expected to be recommendations for further studies related to eHealth-based nutrition education services in Indonesia with the pace of technological development that can be implemented by Indonesia as a developing country.
The inclusion criteria used in this literature review were: (1) include nutritional education interventions in the form of eHealth in experimental studies of obesity prevention and/or management; (2) the age of participants ≥18 years which was an adult age category 16  and/or physical activity. The exclusion criteria in the article search were experimental studies with respondents who had degenerative diseases such as cardiovascular disease, endocrine gland, and neoplasmic disorders as well as bariatric postoperative patients. E-Health intervention studies that did not integrate nutrition education as a major intervention component or part of the study were also excluded from this review. Based on our definition in this systematic review study, eHealth-based nutrition education interventions cover all types of technology channels used as nutritional education messaging media, aimed at assisting respondents in weight loss and anthropometric parameter improvement, supporting a healthy diet, and increased physical activity. Nutritional education interventions delivered using eHealth media may be stand-alone or be part of other multicompetital interventions. This literature study only conducts systematic literature reviews and did not combine data in a meta-analysis.

Study Selection
Based on search results from electronic databases and the manual references the addition, obtained as many as 670 articles relevant to keywords. After eliminating duplication articles, 587 articles were obtained which were then screened to see the suitability of the title and abstract with the purpose of this study and obtained a total of 116 relevant articles. The next stage was to conduct the selection of articles that could be accessed in full as well as a further review of the content of the article based on the criteria of inclusion and obtained the final results of 17 relevant articles. Figure 1. shows a flowchart of one study selection process.
Some studies used 2 to 3 media as channels of communication and activity in the delivery of nutrition education messages and behavioral changes in one intervention period, either given mentoring or independent tasks that must be done by participants in the intervention group. The use of a combination of interpersonal nutrition education methods through telephone/face-to-face personal counseling/group discussion/email with websites/mobile apps/print modules was found in five studies 18, -23,30-31 ; a combination of video conferencing with mobile application/website/ mail/printed module in three studies 20,25,22 ; and a combination of websites and emails in one study 24 . Two studies used≥4 educational media, namely a combination of mobile applications/print media (booklets, leaflets, modules)/email/text messages/counseling by phone or face-to-face 19,31,32 . There were six studies that used only one eHealth medium in its intervention group, namely mobile apps (n=1) 15 ; phone (n=1) 26 ; website (n=3) 17,24,29 and video conferencing (n=1) 27 . Overall, the most widely used media in the intervention group was mobile apps (n=10) 15,[18][19][20][21]23,28,[30][31][32] .
The entire study in this systematic review provides educational materials related to physical activity and sports. Common materials provided in these 17 studies were information on the importance of physical activity, activity recommendations and/or strategies without setting daily and/or weekly (n=10) 15,17,20,22,[24][25][28][29][30]32 , calorie increases, neat calorie increases (n=1) 20 , how to measure exercise intensity (n=2) 20,24 and recommendations for resistance training (n=1) 32 . Seven other studies added information in the form of recommendations for physical activity time of 60 minutes/day (n=1) 31 ; moderate to high intensity exercise in four studies with a time of >180 minutes/week (n =1) 18 and 150 minutes/week (n=3) 21,26-27 and high intensity 60 minutes/week (n=1) 21 ; 10,000 steps/day (n=2) 19,21 ; as well as physical activity 5x/week without old recommendations and intensity of exercise (n=1) 24 . Some of the topics of nutrition education and physical activity materials in these 17 studies were further used as goals and became a tool for self-monitoring of participants through recording food consumption behavior and physical activity. In addition, obesity management methods based on behavior change techniques with the addition of sleep quality regulation interventions did not provide significant results compared to standard methods that emphasized only dietary regulation and physical activity 32 . A comprehensive review of the study can be seen in table 2.

Integration of Theories and Techniques of Behavior Change
Six studies used the theory of behavior change, which uses the integration of two theories (n=2) 18,32 , and four other studies used only one type of theory in their intervention 17,19,29,31 (Table 2). All of these theory-based studies had used a combination of nutritional education and physical activity materials with behavior change techniques and overall had positive results for improvements in anthropometric status (weight loss, % body fat, waist circumference) and eating behavior (diet quality, eating habits) and physical activity (increased exercise time and activity) with assistance (n =5) or independent (n=1) 29 . Studies that did not explicitly report behavioral change theories were 11 studies, with results varying against several outcomes of anthropometric and behavioral status improvements, i.e. significantly positive (n=5) 15,20,24,26,30 and insignificant (n=6) [21][22][23]25,[27][28] .
The whole study used behavior change technique (BCT) strategies in their implementation, but the most widely used in this review were self-monitoring (n=17), goal setting (n=13), and feedback (n=12) through automatic behavioral recording into a mobile application integrated with a weighing tool or pedometer (n=2) 19,21 or manual recording using a notebook/journal (n=2) 26,23 . Other methods of recording behavior were using websites or emails (n=4) 17,22,25,27 and other studies did not explicitly mention recommendations or the use of media for behavioral recording. Variables recorded as selfmonitoring mater for all participants in the majority were weight, type, length of physical activity, and daily food consumption.  between the intervention and control group (p<0.05) with a value of -6.9 cm in the intervention and 1.7cm in the control • There was a significant difference in the weight loss in the 12th week (p<001) with a value of -9.4 Kg in the intervention and -0.6 Kg in the control • In the intervention group, 80% of participants experienced a decrease in weight >5% and 50% of participants >10%.
In the control group, no one >5%. week 52 -4.9 kg (-21.9 to 7.5, p<00.001). • There was a significant difference in median BMI change (1.4 kg/m2, range of -4.5 to 1.1, p<0.001); waist circumference (-3.5 cm, range -23 to 5, p<0.001); %body fat (-2.3kg, range -7.6 to 2.5, p<001) at week 12 and also seen in week 52. • There was a significant difference in eating habits to be healthier (fruit and vegetable consumption, breakfast, reduction of alcohol and sugary drinks and fatty foods) in the 12th week (p<001) and 52nd (p<0.001). There was no significant association of changes in physical activity between baseline and weeks 12 and 52.
friends, educational materials via web-based 15 Huber et al, 2015, US 26 Theory: • -Technique: • Self-monitoring, goal setting, feedback, Intervention: telephone • Calorie deficit • Motivational interviews focus on improving healthier lifestyles, identifying barriers, problem-solving, goal setting • Strategies to achieve deficit calorie and exercise targets Control: Pamphlet • Telephone counseling 1x/ 2 weeks for 3 months making a total of 7x phone calls • Average phone duration of 20 minutes • Monitoring the progress of participants to the targets that have been made and the determination of planning for the next 2 weeks.

Effectiveness of Study Interventions with Short-Term Duration (≤3.5 months / 14 weeks)
Of the 13 studies that had a short-term duration of intervention, 10 studies reported variable weight as anthropometric outcome status with six studies explaining weight loss values ranging from -1.2 kg to -9.4 kg 15,17,[19][20][28][29] and others only explaining the % of significant weight loss without explaining how much the reduction in kilograms (kg) (n=3) 22,25,31 . A total of 5 studies reported BMI outcomes and overall showed significant score reductions in the range of -0.16 to -1.4 kg/m 2 15,17,21,24,31 ; significant waist circumference variables in the three studies 15,19,21 and insignificant in the two studies 26,29 with decreases in the range of -1.4 to -6.9 cm; and variable % body fat was significant in three studies 15,20,21 and not significantly in one study 28 with a decrease in the range of -0.24 to -9.0%. Eating behavior variables were only reported in four studies, related to fruit and vegetable consumption which reported more significantly met the recommendations in the intervention group (n=2) 15,31 and consumption of fat and sweetened beverage foods significantly decreased (n=1) 15 , and there was no significant difference in energy intake (n=1) 28 . Physical activity variables were only reported in two studies with no significant changes in both the intervention group and control 24,31 but still saw an increase in moderate-intensity physical activity of 12 minutes/day in the intervention group versus control 31 . Nutrition education interventions delivered singlehandedly using one eHealth medium provided results as well as multicomponent eHealth media interventions to improve anthropometric outcome variables with different magnitudes of decline in each study with shortterm duration.

Effectiveness of Study Interventions with Long-Term Duration (≥4 months / 16 weeks)
There were 10 studies that evaluated outcomes at several points in the long-term duration category with measurement times at 4 months (n=1) 27 , 6 months (n=4) 18,26,30,32 , 9 months (n=2) 24,31 , 12 months (n=4) 15,18,21,32 , 18 months (n=1) 23 , and 24 months (n=1) 18 . In this long-term study, the use of two nutritional education media in the intervention group (phone application and mentoring) (n=2) 18,30 provided significant results on variable weight loss with an average reduction of ±-3.16 kg in the 6th month and -3.58 kg in the 12th month. But by the 24th month, the difference became insignificant in the average weight loss, diet quality, and physical activity between the intervention group and the control 18 . Interventions with multi-component nutrition education media with ≥3 media (mobile apps, websites, print handouts, and telephone mentoring) also had a significant effect on weight loss outcomes in the 9th month 31 . A study that only used one medium, which was mobile applications 15 and telecoaching via telephone 26 gave significant results on weight loss with values of -4.9 kg and -2.3 kg. Interestingly, inconsistent results were seen in other studies in either the use of one medium or multicomponent. Studies with the use of mobile apps alone without assistance did not have a significant effect on the 12th month 21 and studies with the use of ≥1 media also did not have a significant influence on weight loss variables in the intervention group compared to controls in months 6 and 12 32 .

DISCUSSION
The purpose of interventions that integrate nutritional education materials using eHealth media in this overall review was related to weight management to improve the anthropometric status, changes in eating behavior, and physical activity. The nutritional education methods delivered using eHealth media almost entirely involve modifying behavior changes to facilitate behavior sustainably. By definition of nutrition education itself, it aims to help patients gain the knowledge and skills needed to make changes. Nutrition education and changes in eating habits can have beneficial impacts in many ways including disease control, improved health status, improved quality of life, and decreased health costs 40 .
The results of this study showed that there were two forms of communication channels and activities that are commonly used for the delivery of nutrition education messages in this review, namely interpersonal channels through group discussions and private counseling and mass media channels through the use of the internet (email, websites, applications, video conferences).
EHealth media used as a communication channel for delivery of nutritional education messages in some studies was stand-alone or multicomponent with the most widely used medium in intervention groups being mobile applications. Nutrition education materials containing the topic of the nutritional value of foodstuffs accompanied by daily calorie restriction as well as information on the importance of physical activity and recommendations for physical activity/exercise accompanied by BCT with or without daily targets were the most widely used as nutritional education materials in eHealth-based interventions in this review. Calorie deficits with increased physical activity were a topic recommended by the AHA/ACC/TOS (2013) as a dietary intervention material that can be given to people with more weight and obesity because it has provided strong evidence of successful weight loss 41 .
Delivery of nutritional education materials using eHealth media online provided equally effective results compared to offline direct intervention with the provision of methods based on behavior change techniques and the same duration. This was evident from the three studies in this study showing no significant results on the outcome of anthropometric status (weight loss, BMI), physical activity, and treatment compliance in studies compared eHealth-based methods (online) with face-to-face (offline) which both received obesity management intervention methods accompanied by behavior change techniques in each group (n =4) [22][23]25,27 . In this review, it was also seen that studies using behavior change theory and BCT showed more improvement in expected outcomes rather than studies that did not integrate the theory. In the design of a mHealth-based intervention application, the theory and the BCT selected must be appropriate and interrelated. The ineffectiveness e-ISSN: of an intervention could be due to the theory used not providing a sufficient basis for designing interventions for behavior change that was targeted 42 .
Overall, the most commonly used theory in the study was Social Cognitive Theory (SCT). The SCT theory is a new naming of Learning Theory also developed by Bandura 43 and is a theory for understanding how individuals can translate attitudes and intentions into long-term behavioral changes 12 . In SCT theory, goal setting and self-monitoring are forms of BCT that have been shown to increase self-efficacy and facilitate behavioral change 44 . Another theory of behavior change used in the study was the Transtheoretical Model (TTM) mentioned in two studies 18,31 . TTM theory emphasizes behavioral changes based on five stages that occur sequentially in a short time, ranging from no intention to take action to the continuous stage of behavioral maintenance 45 . Therefore, studies using TTM theory are generally studies with a long duration of time depending on which stage to take as a focus by the researcher.
The integration of nutrition education into eHealth media channels with the provision of nutritional science materials that emphasize modification of eating behavior and physical activity and followed by mentoring both personally and in groups accompanied by BCT that has been designed can have a significant influence on the improvement of anthropometric status and inconsistent results in the outcome of eating behavior and physical activity in obese adults in the short term (≤3.5 months/14 weeks). The use of technology for weight management interventions and healthy lifestyles was less effective on some anthropometric and behavioral status variables if performed in the long term. This was reinforced by previous studies that mentioned that technology-based weight management interventions accompanied by behavioral change engineering components can provide positive results in the short term (3-6 months), but in the long term still provide inconsistent results 46 . Systematic review studies and meta-analyses also showed significantly higher weight loss in interventions combined with dietary modifications and physical activity in behavior-based weight management in obese adults 10 .
Only 1 study 29 reported measuring increased knowledge scores as an outcome variable. Therefore, looking at the definition and scope of nutrition education that was broad 12 , it can be seen from the results of this review that the impact of eHealth media interventions integrated nutrition education was not seen only as limited to the increase in knowledge scores that were short-term but were expected to have more impact on changes in health behaviors. With the interaction between participants and eHealth media used, participants indirectly gained nutritional knowledge through routine learning when using eHealth media. The role of nutrition education in this eHealth-based intervention was to provide nutrition and health information as well as techniques on how to achieve the expected healthy behaviors. The main content in nutrition education was certainly related to individual health, among others, the balance between food consumption and physical activity to the prevention and management of chronic diseases, such as obesity. This will be more effective if it was focused on clearly defined behaviors, practices, or problems. But behavior change itself certainly takes time and must be maintained by overcoming the obstacles 12,[47][48] . It is necessary to design eHealth media that are by the targets so that the effectiveness of interventions can better provide results to expected outcome variables by also involving relevant stakeholders, such as Information Technology/IT developers, health workers in technology planning that can be adjusted to the level of education and socioeconomic and how appropriate methods can be given to prospective users 49 .
The study was a narrow review that specializes only in the assessment of eHealth media that integrates nutrition education in experimental design-based obesity management interventions, not conducting quantitative analysis (meta-analysis) on outcome variables, so the broader discussion may be limited. Apart from this, this study can provide an overview of the recommendations of forms of nutrition education integrated into eHealth media as one form of technology-based intervention on behavior-based obesity management.

CONCLUSION
The integration of nutrition education into eHealth media channels with nutritional science materials that emphasize modification of eating behavior and physical activity, followed by personal or group assistance accompanied by BCT has a significant influence on improving the anthropometric status and inconsistent results in the outcome of eating behavior and physical activity in obese adults in the short term (≤3.5 months/14 weeks). The use of behavioral change theory was designed eHealth-based studies with BCT were more effective at providing significant results toward improvements in several variable outcomes of anthropometric status, eating behavior, and physical activity in short-and long-term studies.