209 JPJO 4 (2) (2019) 209-216 Jurnal Pendidikan Jasmani dan Olahraga http://ejournal.upi.edu/index.php/penjas/index The Effect of Android-based DM Diet Education Program on Eating Behavior of DM Type 2 Patients in Bandung City Health Centres Ides Haeruman Taufik, Lilik Kustiyah, Cesilia Meti Dwiriani Institute Pertanian Bogor , Indonesia Article Info Article History : Received July 2019 Revised August 2019 Accepted August 2019 Available online September 2019 Keywords : Android, Diabetes Mellitus, DM Type 2, Eating Behavior Abstrak Edukasi Diabetes Melitus (DM) adalah pendidikan, pelatihan, pengetahuan dan ket- erampilan dalam mengelola DM. Pendidikan memberikan manfaat penting bagi pen- derita DM, yaitu mendapatkan pengetahuan tentang DM, dan dapat menerapkan infor- masi yang diterima dengan baik dan benar. Tujuan dari penelitian ini adalah untuk menguji pengaruh program edukasi diet DM berbasis Android pada perubahan per- ilaku makan dan kadar glukosa darah. Desain penelitian menggunakan cluster random- ized control trial dengan pre-post intervention study yang dilakukan terhadap 33 pasien kelompok intervensi dan 37 pasien DM kelompok kontrol. Proporsi subyek perempuan lebih tinggi daripada subyek laki-laki antara kelompok perlakuan. Intervensi pendidi- kan gizi dengan media aplikasi secara signifikan meningkatkan skor persepsi tingkat keparahan penyakit DM, ekspektasi pengobatan DM dan praktik kebiasaan makan. Namun, dibandingkan dengan konseling, peningkatan persepsi tingkat keparahan pen- yakit DM secara signifikan berbeda, tetapi untuk persepsi harapan pengobatan DM dan praktik kebiasaan makan tidak berbeda secara signifikan. meskipun peningkatan skor persepsi tingkat keparahan penyakit, harapan pengobatan DM dan kebiasaan makan pada kelompok intervensi lebih tinggi daripada kelompok kontrol. Dengan demikian, penggunaan kedua metode intervensi pendidikan gizi memiliki dampak positif pada peningkatan persepsi keparahan, harapan pengobatan dan praktik kebiasaan makan pasien DM tipe 2. Abstract DM education is education, training, knowledge and skill in managing DM. Education provides important benefits for people with DM to get knowledge about DM and to be able to apply information that is received properly and correctly. The purpose of this study was to examine the effect of Android-based DM diet education program on changes in eating behavior and blood glucose levels. The study design used a cluster randomized control trial with a pre-post intervention study conducted on 33 DM pa- tients in the intervention group and 37 patients in the DM control group. The propor- tion of female subjects was higher than male subjects in treatment groups. Nutrition education intervention with application as the media significantly increased DM dis- ease severity perception score, expectations of DM treatment, and eating habits prac- tice. Compared to counseling, the increase in perception of the severity of DM disease was significantly different, but for DM treatment expectation perceptions and the prac- tice of eating habits were not significantly different. However, the DM treatment ex- pectations, practice of eating habits, and fasting blood sugar checks in the intervention group were higher than the control group. Thus, the use of both nutritional education intervention methods has a positive impact on increasing perceptions of severity, treat- ment expectations, and the practice of eating habits of DM type 2 patients.  Correspondence Address : Jl. Raya Darmaga Kampus IPB Darmaga Bogor, Indonesia E-mail : ides_ht@apps.ipb.ac.id ISSN 2580-071X (online) ISSN 2085-6180 (print) DOI : 10.17509/jpjo.v4i2.19514 210 Ides Haeruman, et.al./ Jurnal Pendidikan Jasmani dan Olahraga 4 (2) (2019) INTRODUCTION According to the data, in 2014, there were 422 million people suffered from diabetes worldwide or the prevalence was 8.5% among adult populations (WHO, 2016). In Indonesia, the data show that there is an in- crease of the diabetes prevalence in Indonesia from 5.7% in 2007 to 8.5% or around 11.2 million in 2018 (Riskesdas, 2018). The data of International Diabetes Federation (2015) present the prediction of the number of diabetes patients in Indonesia is around 10 million. Similar to the situation worldwide, diabetes becomes one of the highest cause of death in Indonesia. The data of Sample Registration Survey in 2014 shows that dia- betes becomes the third highest cause of death in Indo- nesia with the percentage 6.7%, after stroke (21.1%) and heart disease (12.9%). If it is not prevented, this condition would decrease the productivity and cause disability and early death (Kemenkes, 2016). The high number of the DM sufferers has an im- pact on the complication of DM. DM has two complica- tions, including acute complication (hypoglycemia and diabetic ketoacidosis) and chronic complication (macro vascular and micro vascular diseases). The complica- tion of DM could affect our body if the sufferers do not use a proper treatment (Sutedjo, 2010). The complica- tion of DM could be prevented by 4 pillar of manage- ments, including DM education, a good and balance DM diet, the use of a suitable and regular medication, and exercise (Lumenta, 2006). DM education is an education, training, knowledge, and skill in managing and preventing DM (Waspadji, 2009). Education gives an important benefit for DM sufferer including acquiring knowledge related to DM, acquiring ability in acknowledging and facing the symptoms properly, and implementing information properly (PERKENI, 2015) The lack of healthcare infrastructure, the lack of access to the medical experts, and the lack of electronic information systems provided are the causes of unsuc- cessful education of the DM sufferer (Sadowski et al. 2012). In lessening the obstacles on the healthcare ac- cess, the use of new technology, such as smartphone, gives a lot of benefits in the intervention process on the knowledge of DM sufferers (Forjuoh, 2014). The research related to intervention of nutrition education on the diabetes sufferer in Indonesia through the implementation of application in android-based smartphone is limited, even not existed. So far, the ap- plication in android based smartphone only explains about diabetes mellitus in general and there has not been an application that explains and manages the dia- betes mellitus diet specifically. From the above situa- tion, the researcher is interested to test the effect of DM diet education program through android based applica- tion on behavior (knowledge, behavior, and practice) and the blood glucose level of the diabetes type 2 suf- ferers in Bandung Health Centers. The education pro- gram of DM diet is a development of excel based edu- cation program for the nutritionist in conducting nutri- tion consultation. The research of the diet education program is based on android and intended for diabetes sufferers to improve their knowledge on diabetes and food management. METHODS The design of this study was a cluster randomized control trial with pre-post intervention study. The deci- sion of three locations of the research was conducted randomly on 25 health centers that had prolanis pro- gram (Program of chronic disease management) from 71 existed health centers. From 25 selected health cen- ters in the location of the research, three health centers were chosen for the research. The three of health cen- ters were randomized to decide two centers for inter- vention groups and one health center as the control group. The android based application of DM diet was im- plemented to the intervention group and the control group receive counselling from the researcher without the utilization of application. The research was conduct- ed from April – July 2019 for four effective months. This research had been approved by the research ethical commission that involve human subject from the Insti- tut Pertanian Bogor No: 134/IT3.KEPMSMIPB/ SK/2018. Number and Data Collection of Subjects The subject of this research were diabetes type 2 patients in three health centers including Sukajadi Health Centers and Pasundan Health Centers as the in- http://ejournal.upi.edu/index.php/penjas/index DOI : 10.17509/jpjo.v4i2.19514 211 tervention group and Kujangsari Health Center as the control group. Meanwhile, the sample of the study were the member of prolanis program (Program of chronic disease management) and had been screened based on the inclusion and exclusion criteria. The inclusion crite- ria of the research including aged 40 – 85 years, owning android-based smartphone (for intervention group), and were willing to involve in the research by filling in the informed consent. The exclusion criteria are experienc- ing complication that have impacts on DM diet (such as kidney diseases and heart problems), following a thera- py and receiving medicine from another doctor (not the health center doctor), receiving similar intervention from another research, participating in another research, and not willing to obey the research procedure. The minimum number of the subject in this research was calculated by a formula from the research of Widyastuti (2013), as follows: According to the calculation, the number of sub- ject for each group were 30 persons. The calculation was adapted to the condition of the health center, where one prolanis activity included 40 persons. Procedure The subject screening was conducted in three health centers that had been decided through the cluster randomized control trial. The subjects that would fulfil the criteria were decided. The intervention stages in- cludes pre-test (base line), intervention, and post-test (end line). In the base line stage, after the blood sample was taken, the explanation was given to the samples who were willing to join in the research. The explana- tion was conducted by the researcher and the doctor responsible for the health care. After explanation, inter- view and pre-test, by filling in the questionnaire by the sample, were conducted. The questionnaire measured the behavior with Health Belief Model method before intervention. The intervention stage was conducted by giving education of DM diet to the patients of diabetes type 2 patients through android based DM diet program appli- cation for intervention group and PowerPoint Presenta- tion for control group, conducted once by the research- er. The research was helped by the health center offi- cials (at least graduated from diploma degree). The method used in the DM diet education is direct face to face method in the Health Center hall. The technique used was lecturing and practice (for intervention group) that was conducted for 45-60 minutes and discussion for 15-25 minutes. After a month of DM diet intervention education, the repetition of materials and practice by using appli- cation, monitoring, and evaluation of the DM diet edu- cation were conducted. The repetition and monitoring were conducted twice and at the third month, interview and post-test were conducted to measure the diabetes behavior as the last data. Type and Data Collection Method The primer data used in this research includes: the characteristics of the diabetes sufferers (age, sex, edu- cation, number of family member, earning, physical activity, medicine, occupation, period of living with diabetes mellitus, nutritional status/BMI). Data charac- teristics and behavior were collected through question- naire and guided by the researcher and the nutritionist of the health center. Meanwhile, the fasting blood glu- cose level of the sample was conducted with the stand- ardized tools and procedures in the health center labora- tory in Bandung. The secondary data includes the char- acteristics of the health center, the number of prolanis member suffered from DM, and DM diagnosis of prola- nis member. Data Analysis The data process and analysis were conducted by using Microsoft Excel 2013 and SPSS Version 16.0. The research variable were analyzed through descrip- tive analysis, normality test, independent and paired t- test. To see the influence of the DM diet education pro- gram and the level of blood glucose level of diabetes mellitus type 2 patients of each groups analyzed by us- ing independent t-test. If the data were not distributed normally, it would be tested by using Mann Whitney http://ejournal.upi.edu/index.php/penjas/index DOI : 10.17509/jpjo.v4i2.19514 Ides Haeruman, et.al./ Jurnal Pendidikan Jasmani dan Olahraga 4 (2) (2019) 212 test. To find out the behavior differences and the blood glucose level of the DM patients before and after re- ceiving DM diet program for the intervention group and for the control group was analyzed by paired t-test. If the data distribution was not normal, the data would be analyzed by Wilcoxon. RESULT The subject of this research were the patients in the health centers in Bandung, including Pasundan Health Center, Sukajadi Health Center, and Kujangsari Health Center. The result of t-test shows that there is no signifi- cant differences between intervention and control group whether in the severity of the DM, the benefits of medi- cation, and the subject diet practice for intervention and control group before intervention. Spearman test was conducted to find out the rela- tionship between the characteristics of the subject with the severity of DM, the benefits of medication, and diet practice before intervention. Table 2 shows that the pe- riod of the patient suffered from DM has a significant positive relationship with the medication expectancy of the subject before intervention (r=0.26, p<0.05). Sex, age, education, activity, occupation, income, and BMI did not influence the severity of DM, medication expectancy, and diet practice. Table 3 shows the intervention of DM diet educa- tion, in general, indicates an improvement on be- havior (perception of DM severity and medication benefits) from baseline to end line. The distribu- tion of subjects related to the severity of DM when pre-test (baseline) were mostly on the medium category for intervention group (51.5%) and con- trol group (45.9%). The result of t-test between intervention group and control group shows that there is a significant differences between pre-test score of the intervention and control group (Table 3). Most of the subject were in the medium, high cat- egory related to DM medication expectancy per- ception before intervention whether for interven- tion group or control group (Table 4). The result of t-test in pre-test (baseline), and post-test (end line) between the intervention groups shows that nutritional education, through android application or counseling, gave a positive impact on the score of perception of medication expectancy for DM. Perception of DM medication expectancy in the intervention group increased significantly in the end line data collection if compared to the base- line data. It shows that the repetition of exposure could give significant impacts on the higher in- crease. Nutritional education would be better to be conducted continuously and repetitively to have a better attitude and behavior (Khomsan et al. 2007). http://ejournal.upi.edu/index.php/penjas/index Table 1. Distr ibution of Subjects based on Individual Characteristics Severity of DM Medication Diet Practice (pre-test) (pre-test) (pre-test) Sex Male 6.0±1.1 15.7±1.7 1.5±0.5 Female 6.2±1.6 15.6±1.3 1.6±0.5 p=0.51, r=0.08 p=0.92, r=-0.01 p=0.3, r=-0.13 Age <45 7.0± - 14.0± - 2.0± - 45 - 54 7.0±0.7 16.2±1.3 1.6±0.5 55 – 64 5.8± 15.6±1.4 1.5±0.5 65 – 74 6.2± 15.7±1.6 1.7±0.5 >75 5.3± 15.5±1.4 1.3±0.5 p=0.49, r=-0.08 p=0.94, r=-0.01 p=0.93, r=0.02 Education Do not Graduated from Primary School 6.0±1.0 16.3±1.5 0.3±0.6 Primary School 5.3±2.3 15.1±1.6 0.2±0.4 Junior High School 6.2±0.9 16.3±1.0 0.4±0.5 Senior High School 5.9±1.5 15.4±1.4 0.5±0.5 Higher Education 6.5±1.1 15.7±1.7 0.4±0.5 p=0.26, r=0.14 p=0.62, r=-0.06 p=0.26, r=0.14 Activity Light 5.8±1.5 15.6±1.6 0.4±0.5 Moderate 6.2±1.5 15.9±1.6 0.4±0.5 Heavy 5.9±1.1 14.9±1.5 0.4±.52 p=0.77, r=0.04 p=0.77, r=-0.04 p=0.77, r=-0.04 Occupation House Wife 5.9±1.6 15.8±1.4 0.4±0.5 Retirement 6.1±1.1 15.1±1.6 0.5±0.5 Private Employee 6.4±1.1 16.0±1.6 0.4±0.5 p=0.60, r=0.06 p=0.41, r=-0.10 p=0.57, r=0.07 Income < Regional Minimum Income 6.0±1.4 15.6±1.5 0.4±0.5 ≥ Regional Minimum Income 6.0±1.5 15.7±1.4 0.4±0.5 p=0.96, r=0.01 p=0.84, r=0.02 p=0.69, r=0.05 Period of Suffered from DM <10 years 6.1±1.3 15.3±1.4 0.4±0.5 ≥ 10 years 5.9±1.6 16.0±1.4 0.4±0.5 p=0.71, r=-0.05 p=0.031, r=0.26 p=0.81, r=0.03 BMI Thin 5.3±3.8 15.3±0.6 - ± - Normal 6.1±1.3 15.5±1.4 0.5±0.5 Overweight 6.0±1.4 15.9±1.4 0.3±0.5 Obese 5.9±1.1 16.0±2.0 0.8±0.5 p=0.45, r=-0.09 p=0.28, r=0.13 p=0.49, r=0.08 1 significant if p<0.05 with Mann-Whitney test DOI : 10.17509/jpjo.v4i2.19514 Ides Haeruman, et.al./ Jurnal Pendidikan Jasmani dan Olahraga 4 (2) (2019) 213 http://ejournal.upi.edu/index.php/penjas/index Table 2. Relationship between Subject Char acter istics and the Sever ity of DM, medication expectancy, and fasting diet practice before intervention. Characteristics Severity of DM Medication Expectancy Diet Practice (pre-test) (pre-test) (pre-test) Sex Male 6.0±1.1 15.7±1.7 1.5±0.5 Female 6.2±1.6 15.6±1.3 1.6±0.5 p=0.51, r=0.08 p=0.92, r=-0.01 p=0.3, r=-0.13 Age <45 7.0± - 14.0± - 2.0± - 45 - 54 7.0±0.7 16.2±1.3 1.6±0.5 55 – 64 5.8± 15.6±1.4 1.5±0.5 65 – 74 6.2± 15.7±1.6 1.7±0.5 >75 5.3± 15.5±1.4 1.3±0.5 p=0.49, r=-0.08 p=0.94, r=-0.01 p=0.93, r=0.02 Education Do not Graduated from Primary School 6.0±1.0 16.3±1.5 0.3±0.6 Primary School 5.3±2.3 15.1±1.6 0.2±0.4 Junior High School 6.2±0.9 16.3±1.0 0.4±0.5 Senior High School 5.9±1.5 15.4±1.4 0.5±0.5 Higher Education 6.5±1.1 15.7±1.7 0.4±0.5 p=0.26, r=0.14 p=0.62, r=-0.06 p=0.26, r=0.14 Activity Light 5.8±1.5 15.6±1.6 0.4±0.5 Moderate 6.2±1.5 15.9±1.6 0.4±0.5 Heavy 5.9±1.1 14.9±1.5 0.4±.52 p=0.77, r=0.04 p=0.77, r=-0.04 p=0.77, r=-0.04 Occupation House Wife 5.9±1.6 15.8±1.4 0.4±0.5 Retirement 6.1±1.1 15.1±1.6 0.5±0.5 Private Employee 6.4±1.1 16.0±1.6 0.4±0.5 p=0.60, r=0.06 p=0.41, r=-0.10 p=0.57, r=0.07 Income < Regional Minimum Income 6.0±1.4 15.6±1.5 0.4±0.5 ≥ Regional Minimum Income 6.0±1.5 15.7±1.4 0.4±0.5 p=0.96, r=0.01 p=0.84, r=0.02 p=0.69, r=0.05 Period of Suffered from DM <10 years 6.1±1.3 15.3±1.4 0.4±0.5 ≥ 10 years 5.9±1.6 16.0±1.4 0.4±0.5 p=0.71, r=-0.05 p=0.031, r=0.26 p=0.81, r=0.03 BMI Thin 5.3±3.8 15.3±0.6 - ± - Normal 6.1±1.3 15.5±1.4 0.5±0.5 Overweight 6.0±1.4 15.9±1.4 0.3±0.5 Obese 5.9±1.1 16.0±2.0 0.8±0.5 p=0.45, r=-0.09 p=0.28, r=0.13 p=0.49, r=0.08 1 significant if p<0.05 with Spearman test DOI : 10.17509/jpjo.v4i2.19514 Ides Haeruman, et.al./ Jurnal Pendidikan Jasmani dan Olahraga 4 (2) (2019) 214 The distribution of subject based on the practice of eating habit when pre-test (baseline) most of them were in the disobedient category on DM diet for intervention group (51.5%) and for control group (64.9%). The re- sult of t-test for intervention and control group shows that there was no significant differences between the pre-test score of intervention group and control group (Table 5). DISSCUSSION Most of the research subject were the diabetes pa- tient involved in prolanis group in three heath centers in Bandung. The distribution of subject related to the se- verity of DM in pre-test (baseline) were in the medium category for both intervention group (51.5%) and con- trol group (45.9%). The result of t-test between inter- vention and control group shows that there is a signifi- cant difference between the intervention group and con- trol group. After intervention, there is a significant knowledge improvement for both of intervention groups. There is score improvement related to percep- tion of the DM severity for 1-1.2 points. The score of the DM severity perception in post-test is different sig- nificantly between the intervention groups. The score of improvement in perception of DM severity score, the intervention group is higher 0.2 points in comparison with the control group, with most of the subjects are in the high category (93.9%) while in the control group only 37.8% who are in the high category. The education media is a helping tool to deliver nutritional and health information easier. The media could prevent a misper- ception, give brief information, and easier to give un- derstanding (Arsyad 2009; Fitriani 2011). Similar to the DM severity perception, in general, there is a positive improvement on the DM medication expectancy from baseline to end line. However, some statements were responded by the subject inaccurately that causes the post-test score is lower that the pre-test score. One of those is the low understanding on DM medication expectancy that only through medicine, not holistically in five pillars of DM management and have boredom in doing DM diet. It might be because the sub- ject experience a boredom in medication and have an inaccurate perception in DM medication, where the medication only relies on medicine and perceives if http://ejournal.upi.edu/index.php/penjas/index Table 3. Subject Distr ibution based on sever ity per cep- tion of DM Severity of DM Interven- tion Control Total p value n (%) n (%) n (%) Pre-test Low 2 (6.1) 6 (16.2) 8 (11.4) Medium 17(51.5) 17 (45.9) 34 (48.6) High 14(42.4) 14 (37.8) 28 (40.0) Score (average±SD) 6.2±1.3 5.9±1.5 6.0±1.5 0.3621 Post-test Low 1 (3.0) 1 (16.2) 2 (2.9) Medium 1(3.0) 12 (45.9) 13 (18.6) High 31(93.9) 24 (37.8) 55 (78.6) Score (average±SD) 7.6±0.8 6.9±1.1 7.2±1.0 0.0021 Delta (Post-test – Pre-test) 1.42 1 1.2 0.043 p value 0.0002 0.0002 0.0002 1 significant if p<0.05 with t-test intervention and control group Table 4. Distr ibution of Subject based on the per ception Medication Expectancy of DM Interven- Control Total p value n (%) n (%) n (%) Pre-test Low 0 (0.0) 0 (0.0) 0 (0.0) Medium 11(33.3) 22 (59.5) 33 (47.1) High 22 (66.7) 15 (40.5) 37 (52.9) Score (average±SD) 16.0±1.4 15.4±1.5 15.7±1.5 0.0621 Post-test Low 0 (0.0) 0 (0.0) 0 (0.0) Medium 2 (6.1) 7(18.9) 9(12.9) High 31 (93.9) 30(81.1) 61(87.1) Score (average±SD) 17.8±1.4 16.8±1.6 17.3±1.6 0.0051 Delta (Post-test – Pre-test) 1.8 1.4 1.6 0.230 p value 0.0002 0.0002 0.0002 1 significant if p<0.05 with t-test between the intervention nd control group 2 significant if p<0.05 with paired t-test (Wilcoxon) Table 5. Distr ibution of Subject based on Pr actice of Practice of Eat- ing Habit Intervention Control Total p value n (%) n (%) n (%) Pre-test Disobedient 17 (51.5) 24 (64.9) 41 (58.6) Obedient 16 (48.5) 13 (35.1) 29 (41.4) Score (average±SD) 0.5±0.5 0.4±0.5 0.4±0.5 0.2611 Post-test Disobedient 12 (36.4) 21 (56.8) 33(47.1) Obedient 21 (63.6) 16 (43.2) 37(52.9) Score (average±SD) 0.6±0.5 0.4±0.5 0.5±0.5 0.091 Delta (Post-test – Pre-test) 0.1 0 0 0.359 p value 0.0252 0.0832 0.052 1 significant if p<0.05 with t-test between the intervention and control group DOI : 10.17509/jpjo.v4i2.19514 Ides Haeruman, et.al./ Jurnal Pendidikan Jasmani dan Olahraga 4 (2) (2019) 215 medication only could heal DM totally. The low num- ber of subject 60.6% on the subject from intervention group and 35.1% from control group who answered that medicine only could control the blood glucose level was caused by the inaccurate understanding in manag- ing DM where the patients really rely on medication. Most of the subjects, before intervention, were in the medium, high category for medication expectancy of DM for both intervention and control groups (Table 4). Similar to the perception of the severity of DM, the significant result of pre-test (baseline) and post-test (end line) between intervention group shows that nutri- tion educations, through android application media or counseling, give a positive impact on the improvement of DM medication expectancy score. However, it can- not be seen which intervention method that is better in giving contribution on the improvement of the DM medication expectancy perception. The improvement of the subject behavior related to the perception of medi- cation expectancy shows a significant result for both intervention group. Similar to the severity perception of DM, the significant result of t-test in pre-test (baseline) and post-test (end line) between intervention groups shows that nutrition educations, through android appli- cation or counseling, give positive impacts on the im- provement of score of DM medication expectancy per- ception. However, it cannot be seen which method that has a better contribution on the improvement of the per- ception of DM medication expectancy. The improve- ment of subject attitude related to the perception of DM medication expectancy shows significant result for both intervention groups. The measurement of eating habit practice in post- test (end line) showed a significant improvement 0.1 point for intervention group, while for the control group did not have score improvement. The result of the t-test between intervention groups showed a significant dif- ference between intervention groups for eating habit practice score. Although nutrition education interven- tion through application media gives a little better im- provement compared to counseling media, but, statisti- cally the difference was not significant (p=0.359). The significant difference in pre-test (baseline) and post-test (end line) between intervention group shows nutrition education through android application gave a positive impact on the improvement of eating habit practice, while in the control group statistically did not have changes or intervention on control group did not give any influences. Although intervention method on appli- cation group give a good result, however it cannot be seen which method that give better contribution on the eating habit practice. It is related to the statement of Andragogy behav- ior changes reaction caused by intervention (counseling) on adults could be assessed after one month. The changes on eating behavior is the improve- ment of knowledge, attitude, and practice in the diet management that is needed to control blood glucose level (Thaha 2003). In the process of behavior changes, it should be considered that the knowledge change to attitude and to practice is not a linear line. The change of knowledge to attitude is affected by perception related to the problem and the changes. If the attitude has changed, it is the predisposition for behavior changes (Ngatimin 2005). After intervention, there was an improvement on eating habit score for intervention group while for the control group statistically did not experience changes. This result is relevant with the research that showed intervention had a positive impact on some clinical re- sults and self-efficacy. However, the clinical manage- ment technology should be made friendlier for the users before a larger trial phase II is conducted (Faridi et al. 2008; Quinn et al. (2008). CONCLUSION The nutritional education media based on applica- tion with DM diet material through health believe mod- el successfully improved the perception score of DM severity (22.9%), perception of DM medication expec- tancy (11.3%), eating habit practice (20%), and the de- crease of fasting blood glucose level (0.1%) from the baseline. In compare with the control group, the im- provement of perception of DM severity score is differ- ent significantly. However, in the DM medication vari- able, the practice of eating habit, and the fasting blood glucose level check are not different significantly. Therefore, it indicates that DM diet intervention gives positive impacts on the score improvement of all re- search variables. The nutrition education with applica- tion media should be conducted continuously to have a http://ejournal.upi.edu/index.php/penjas/index DOI : 10.17509/jpjo.v4i2.19514 Ides Haeruman, et.al./ Jurnal Pendidikan Jasmani dan Olahraga 4 (2) (2019) 216 change on behavior and eating habit practice that would give benefits in controlling fasting blood glucose level of diabetes patients. REFERENCES Arsyad A. 2009. Media Pembelajaran. Jakarta (ID): Raja Rafindo Persada. Faridi Z, Liberti L, Shuval K, Northrup V, Ali A, Katz DL. 2008. Evaluating the impact of mobile tele- phone technology on type 2 diabetic patients’ self- management: the NICHE pilot study. J Eval Clin Pract. 14(3): 465–469 Fitriani S. 2011. Promosi Kesehatan. Yogyakarta (ID): Graha Ilmu. Forjuoh SN, Bolin JN, Huber JC, John CHJ, Ann VM, Omolola AE, Janet HW, Dawn BS, Anne R. 2014. Behavioural and technological interventions target- ing glycemic control in a racially/ethnically diverse population: a randomized controlled trial. J BMC Public Health. 14: 71. [KEMENKES RI] Kementerian Kesehatan Republik Indonesia. 2016. Mari Kita Cegah Diabetes Dengan Cerdik. [Internet]. [diunduh 2018 Mei 17]. Tersedia dari: http://www.depkes. go.id/article/ print/16040700002/menkes-mari-kita-cegah-diabetes -dengan-cerdik. html. _____________________________________________ _______. 2018. Riset Kesehatan Dasar 2018. Jakarta (ID): Kementrian Kesehatan RI. _____________________________________________ ______. 2014. Info Datin (Pusat data dan informasi Kementerian Kesehatan RI) Situasi dan analisis die- bates. Jakarta (ID): Kementrian Kesehatan RI. Khomsan A. 2000. Teknik Pengukuran Pengetahuan Gizi. Bogor. J Institut Pertanian Bogor. 30 – 34. Lumenta N. 2006. Kenali Jenis Penyakit dan Penyem- buhannya: Manajemen Hidup Sehat. Jakarta (ID): PT. Elex Media Komputindo, Gramedia. Ngatimin HMR. 2005. Sari dan Aplikasi Ilmu Perilaku Kesehatan. Makasar (ID): Yayasan “PK-3”. [PERKENI] Perkumpulan Endokrinologi Indonesia. 2015. Konsensus Pengelolaan dan Pencegahan Dia- betes Melitus Tipe 2 Di Indonesia. Jakarta (ID). PB PERKENI. Quinn CC, Clough SS, Minor JM, Lender D, Okafor MC, Gruber-Baldini A. 2008. WellDoc mobile dia- betes management randomized controlled trial: change in clinical and behavioural outcomes and patient and physician satisfaction. J Diabetes Tech- nol Ther. 10(3): 160–168. Sadowski D, Devlin M, Hussain A. 2012: Diabetes self- management activities for Latinos living in non- metropolitan rural communities: a snapshot of an underserved rural state. J Immigr Minor Health, 14 (6): 990–998. Sutedjo. 2010. 5 Strategi Penderita Diabetes Melitus Berusia Panjang. Yogyakarta (ID): Kanisius. Thaha R. 2003. Pengaruh Komunikasi Interpersonal terhadap pemanfaatan Pelayanan Bagi Perawatan bayi dan Balita Sakit[tesis]. Bandung (ID): Progran Studi Promosi Kesehatan Masyarakat Fakultas Kesehatan Masyarakat Universitas Padjajaran. Waspadji S. Soegondo S, Soewondo P, Subekti I. 2007. Diabetes Melitus: Mekanisme dasar dan pengel- olaannya yang rasional. Jakarta (ID): Balai Penerbi- tan FKUI. [WHO] World Health Organization. 2016. Global Re- port On Diabetes. WHO [internet]. [diunduh 2018 Agustus 9]. Tersedia dari pada: http://apps.who.int / iris/bitstream/ handle/10665/204871/9789241565257_eng.pdf;jsess ionid=E33166BB7F3A0251865816F9DDE65D2D? sequence=1 Widyastuti D. 2013. Pengaruh Konseling Gizi Ber- dasarkan Teori Health Belief Model Terhadap Pengetahuan Dan Kepatuhan Diet Pada Pasien DM Tipe 2 Di Rawat Jalan RSUP Dr Hasan Sadikin Ban- dung[Tesis]. Bandung (ID): Progran Studi Promosi Kesehatan Masyarakat Fakultas Kesehatan Masyara- kat Universitas Padjajaran http://ejournal.upi.edu/index.php/penjas/index DOI : 10.17509/jpjo.v4i2.19514 Ides Haeruman, et.al./ Jurnal Pendidikan Jasmani dan Olahraga 4 (2) (2019)