|Year : 2020 | Volume
| Issue : 2 | Page : 90-100
Impact of noncommunicable disease text messages delivered via an app in preventing and managing lifestyle diseases: Results of the “myArogya” worksite-based effectiveness study from India
Harish Ranjani1, Sharma Nitika1, Ranjit Mohan Anjana2, Sandhya Ramalingam3, Viswanathan Mohan2, Nalini Saligram3
1 Department of Translational Research, Madras Diabetes Research Foundation, Chennai, India
2 Department of Diabetology, Madras Diabetes Research Foundation and Dr Mohan’s Diabetes Specialities Centre, WHO Collaborating Centre for Noncommunicable Diseases Prevention and Control, IDF Centre for Education, Gopalapuram, Chennai, India
3 Arogya World India Trust, Bangalore, India
|Date of Submission||02-Feb-2019|
|Date of Decision||20-May-2019|
|Date of Acceptance||06-Jun-2019|
|Date of Web Publication||24-Jun-2020|
Dr. Harish Ranjani
Madras Diabetes Research Foundation (MDRF), No. 4, Conran Smith Road, Gopalapuram, Chennai 600 086, Tamil Nadu.
Source of Support: None, Conflict of Interest: None
Background: Smartphones provide an opportunity for preventing lifestyle diseases through mobile applications. Purpose: The aim of this study was to investigate the effectiveness of a mobile phone application (app) in making lifestyle changes in the community in individuals across glycemic categories. Materials and Methods: Adult participants (n = 674) across eight worksites were randomized into an intervention group that downloaded the “myArogya” app and received noncommunicable disease prevention messages twice a week for 6 months in the form of modules and into a control group who did not receive the app. Both groups attended a talk on diabetes prevention and management at baseline. Clinical and biochemical parameters were screened at baseline and postintervention. Eating and behavior scores were computed based on a set of questions on lifestyle habits. The primary outcomes were compared across study arms and within glycemic categories. Results: Intervention group participants with prediabetes showed a greater percent change in HbA1c levels (−1.9%) and blood pressure levels (systolic −2.7% and diastolic −3.1%), and a significant increase in exercisers (8%) compared to controls. Both intervention and control group participants reported that they had quit smoking and showed similar increase in moderate-to-vigorous physical activity, mean eating score, and total behavior score. A similar percentage of intervention (47%) and control group (48%) participants achieved two or more prescribed goals. Conclusion: “myArogya” app helped to reduce blood pressure and smoking behaviors, adopt healthy eating, and improve physical activity level among the study participants. Prediabetes and diabetes participants were more inclined toward making positive lifestyle changes as compared to participants in normal glucose tolerance category. However, similar number of participants met the study goals across groups.
Keywords: BP, diabetes, HbA1c, lifestyle goals and behavior, myArogya app
|How to cite this article:|
Ranjani H, Nitika S, Anjana RM, Ramalingam S, Mohan V, Saligram N. Impact of noncommunicable disease text messages delivered via an app in preventing and managing lifestyle diseases: Results of the “myArogya” worksite-based effectiveness study from India. J Diabetol 2020;11:90-100
|How to cite this URL:|
Ranjani H, Nitika S, Anjana RM, Ramalingam S, Mohan V, Saligram N. Impact of noncommunicable disease text messages delivered via an app in preventing and managing lifestyle diseases: Results of the “myArogya” worksite-based effectiveness study from India. J Diabetol [serial online] 2020 [cited 2021 Jan 25];11:90-100. Available from: https://www.journalofdiabetology.org/text.asp?2020/11/2/90/287612
| Introduction|| |
Noncommunicable diseases (NCDs) are now a much larger cause of morbidity and mortality than communicable diseases. Globally, 415 million people now have diabetes, and this number is estimated to reach 642 million by 2040. Importantly, health care costs continue to increase with 12% of global health expenditure dedicated to the treatment of diabetes and its complications. Today 75% of people with diabetes are estimated to live in low- and middle-income countries, where the people end up spending over 70% of their incomes on health thus further pushing people below the poverty line. Despite lower overweight and obesity over 72 million individuals have been diagnosed with diabetes in India, thus making it a significant public health problem.,, According to the Indian Council of Medical Research–India Diabetes study, it is estimated that the overall prevalence of diabetes in India is 7.3% and the prevalence of prediabetes is 10.3%.
Diabetes self-management education is a key component of care for all people with diabetes and prediabetes., Several randomized controlled trials (RCTs) have shown that lifestyle modification can reduce conversion from prediabetes to type 2 diabetes (T2DM). Communication technologies have changed dramatically; thanks to mobile phone technologies. This in turn has led to the expansion of mobile-health (mHealth), which refers to the use of mobile computing and communication technologies in public health care.,,,,,,
Arogya World, a global nongovernmental nonprofit organization earlier, carried out one of the largest diabetes prevention “mHealth” programs—mDiabetes—with the help of Nokia Life and transmitted more than 56 million preventive text messages to the consumers in 2012. The results showed promising trends in behavior change with an 11% increase in daily exercise, 15% increase in the intake of 2–3 servings of fruits a day, and an 8% increase in 2–3 servings of vegetables per day, and these results were statistically significant.
This study aimed at investigating the effectiveness of the Arogya World mobile phone application (myArogya app) in the primary prevention of diabetes through lifestyle changes carried out at eight Arogya World Healthy Workplaces, based at Bengaluru, in Southern India (CTRI/2015/12/006469).
| Materials and Methods|| |
In this study, we tested the effectiveness of the Arogya World app in the primary prevention of diabetes in participants with varying stages of glycemia (normal glucose tolerance [NGT], prediabetes [Pre diab], self-reported diabetes mellitus [DM], and newly diagnosed diabetes [NDD]).
Diabetes is a chronic condition that occurs when the body cannot produce enough insulin or cannot use insulin, and is diagnosed by observing raised levels of glucose in the blood. “Type 1” diabetes is caused by an autoimmune reaction, in which the body’s defense system attacks the insulin-producing beta cells in the pancreas. As a result, the body can no longer produce the insulin it needs. In T2DM, the body is able to produce insulin but becomes resistant so that the insulin is ineffective. Over time, insulin levels may subsequently become insufficient. Both the insulin resistance and deficiency lead to high blood glucose levels.
People with raised blood glucose levels that are not high enough for a diagnosis of diabetes are said to have impaired glucose tolerance or impaired fasting glucose. These conditions are sometimes called “pre-diabetes.”
- (1) Weight loss—decrease in 3kg of weight and/or decrease in waist circumference by 3cm at the end of intervention (6 months)
- (2) Blood pressure (BP)—decrease of 5mm Hg in systolic BP and 2mm Hg of diastolic pressure
- (3) Lipids—increase in high-density lipoprotein cholesterol by 2 mg/dL
- (4) HbA1c—decrease by 0.2%
The primary objective was to compare the proportion of participants who achieve at least 2 of the 4 goals with 6% clinically significant difference in risk factor reduction between intervention and control groups.
The secondary objectives were:
- (a) to study the effectiveness of the app (measured by way of goals achieved) in each glycemic group, and
- (b) to study how many lifestyle goals (150 minutes or more of physical activity (PA) and/or adoption of healthy eating behaviors) could be achieved across and within each glycemic group.
Recruitment of participants and eligibility criteria
The recruitment of the participants began with an introductory talk to the employees of the myArogya healthy workplaces about diabetes, the importance of the study, and the risks involved. All those who were interested to participate had to undergo a two-phase screening process to assess their eligibility for the study.
Eligibility criteria were as follows: no major illness such chronic liver disease, kidney disease, or cancer; no cognitive impairment; no severe depression or mental imbalance; no physical disability that would prevent regular PA; no participation in another trial; aged 20–65 years; and ownership of a smartphone and ability to read and understand mobile phone messages in English. Participants with type 1 diabetes were not included in the study.
Sample size calculations
Using the earlier primary prevention trial, Diabetes Community Lifestyle Improvement program as the reference (8), in this study we expected that 16% of the intervention group and 10% in control group would achieve 2 of the 4 goals. For 80% power, the minimum sample size required was 492 in each group using OpenEpi online sample size calculator. Thus, the total sample size was 1850 participants after accounting for 20% loss to follow-up and a design effect of 1.5 (accounting for clustering effect at worksites) with a power of 80% (i.e., 925 people each in two groups).
The prescreening survey questionnaire was answered by 4065 participants across the 8 worksites, of which 155 were ineligible for the study. The remaining participants (3910) were invited for a baseline screening camp. Of whom, 2558 (63%) participants attended and were randomized to either intervention (n = 1281) or control group (n = 1277). At postintervention, 674 participants completed the study, 322 in the intervention group and 352 in the control group [Figure 1].
Study design and timeline [[Figure 2]]
Phase 1 screening: After the introductory talk on diabetes, link to a short online questionnaire was sent to all the employees with the help of their human resources department. The questionnaire had questions about sociodemographics and assessment of risk for diabetes using the Indian Diabetes Risk Score (IDRS). All those who met the eligibility criteria and had a high risk for diabetes or already had diabetes were invited for the phase 2 screening through an e-invite sent to their e-mail ids.
Phase 2 screening: All the participants were briefed regarding the details of the study. A written informed consent was obtained and a copy was given to the participants, following which a unique identification number was created for all those who attended the phase 2 screening. The screening included a blood draw in a non-fasting state to measure the blood sugar and lipids (fats) in the blood. Participants also gave their urine samples. In addition, BP and anthropometric measurements (height, weight, and waist circumference) were recorded during this visit by trained professionals from a well-known diagnostic laboratory chain and their reference ranges were used. All the readings were taken twice and recorded in a predesigned format. Reports were sent to the personal e-mail ids of all the participants. At every stage of screening, all patient safety and data privacy regulations were adhered to. Individuals who had abnormal test results, e.g., elevated glucose, BP, or lipids, were referred to their primary care physician for clinical follow-up.
Lipid profile tests were performed using spectrophotometry method. The sample was collected in vacutainer serum separator tube. The blood sample was then centrifuged, allowing the clear serum to be removed for testing. HbA1c test was performed using immunoturbidimetry method. The sample was collected in vacutainer ethylenediaminetetraacetic acid tubes. Specimen bags were used to safely transfer specimens and paperwork. The zipper bag holds specimen and prevents spills. And the outer pouch keeps paperwork separate and secure. The specimen bags were packed in a thermocol box and were safely transferred to the centralized lab for testing.
Randomization of sample: Participants were randomly assigned to either the intervention or control group using a computer-generated randomization grid and were further randomized based on their glycemic status into normal glucose tolerant, prediabetes, and diabetes groups.
Participants were informed about which group they belonged to through e-mail and were then followed up with phone calls. Information regarding the intervention group was shared with the app developer (ClickMedix) to help them roll out the app only to the intervention group participants. The details of the app are provided in Appendix 1.
Online links to complete the behavior change questionnaire were e-mailed to the participants. All the data sets were merged after collecting the behavior survey data.
The eligible participants once assigned to the intervention group received the notification to download the “myArogya” app. The study team helped the participants with the onboarding of the app. They were asked to register and choose one of the four message modules. They had to log in to the app and create a profile. Following this, they received messages two times a week for 6 months. In the case of combination module, the messages were more frequent. Messages were also sent via e-mail. The app also contained a food and activity tracker that helped participants to plan their meals and make appropriate lifestyle changes. The control group did not receive the app at baseline but were given the app after completion of the study. Health checkup reports and the mobile app were provided free of charge to study participants. They were then invited for follow-up testing 6 months after the intervention (postintervention phase).
Behavior survey scoring
To compare the lifestyle behavior changes over time, a composite healthy behavior change score was computed. Each instance of pre/post behavior that was healthy was assigned a numeric score of 1 whereas unhealthy behavior was assigned a score of 0. Sixteen questions were considered for analysis of participant awareness about diabetes and heart disease, PA practices, and dietary guidelines. The scores for each of the four behaviors were then summed, generating a composite score at each phase of the study and these scores were then compared. Separate scores for PA behavior and eating habits were also calculated.
The analysis was conducted in SPSS version 16.0. The study was designed to provide 80% power to detect a 6% difference in achieving 2 of the 4 lifestyle goals between both the study groups. Intervention adherence was assessed within (pre/postintervention) and between intervention and control groups by evaluating the following:
Biochemical parameters and anthropometric measurements as mean values and % changes compared by Student’s t test.
Changes in weight, waist circumference, HbA1c, and BP as a composite score for cardiometabolic risk factor reduction was compared using Z scores and t test.
Changes in lifestyle and behavior as a composite score compared using Z score and t test.
Differences in parameters studied within the study groups are presented as means and standard deviation, whereas across the groups they are presented as percent change. The formula used to calculate percent change was [(mean at postintervention − mean at baseline)/(mean at baseline)] ×100
For analysis, self-reported DM and NDD (defined as participants who were not known cases of diabetes but had HbA1c levels more than or equal to 6.5%) were merged together.
The Institutional Ethics Committee of Madras Diabetes Research Foundation approved the study (Reg No:ECR/194/Inst/TN/2O13).
| Results|| |
A total of 2558 participants attended the baseline screening camp and were randomized to the intervention (n = 1281) and control (n = 1277) groups. A total of 674 participants (26%) attended the postintervention screening camps (intervention group [n = 322] and control group (n = 352)].
Majority of the study participants were males (75%) and the mean age was 31 ± 6.4 years. Of the participants, 76% were in the age group of 20–35 years.
[Table 1] shows that participants across both groups had a high risk for diabetes based on their IDRS. Majority (62%) of the participants had NGT, 28% had prediabetes, and 10% had diabetes, which includes self-reported DM (5.3%) and 4% NDD.
|Table 1: Indian Diabetes Risk Score and distribution of participants across the glycemic categories|
Click here to view
Anthropometric, clinical, and biochemical parameters
Normal glucose tolerance
The NGT participants in the intervention group showed a statistically significant reduction in mean systolic (123.1 ± 12.2 to 119.6 ± 13.5) and mean diastolic BP (76.5 ± 9.9 to 74.4 ± 9.8) and an increase in waist circumference (88.7 ± 9.2 to 89.8 ± 9.1); P = 0.01) and HbA1c levels (5.3 ± 0.2 to 5.4 ± 0.3) postintervention [Table 2].
|Table 2: Comparison of mean values of anthropometric and biochemical parameters within the intervention and control groups|
Click here to view
Among the control group participants, there was significant increase in the weight (70.3 ± 11.9 to 70.9 ± 11.7) and waist circumference (88 ± 9.2 to 89.5 ± 9.5) and a decrease in mean systolic (124.3 ± 13 to 121.4 ± 13.8) and diastolic BP (77.3 ± 10.2 to 75.9 ± 10) [Table 2].
Comparison across the intervention and control groups showed that the intervention group had a better percent reduction in systolic and diastolic BP (−2.7, −2.3) as compared to control group (−2.1, −1.2) though the results were not significant [Table 2].
Among the prediabetes participants in the intervention group, the mean systolic and diastolic BP reduced from 125.2 ± 10.8 to 121.5 ± 12.4 (P = 0.002) and 78 ± 8.4 to 75.4 ± 9.6 (P = 0.002), respectively and mean HbA1c from 5.9 ± 0.2 to 5.8 ± 0.3 (P < 0.0001) [Table 2]
The control group also showed similar significant reductions in these parameters. [Table 2]
Comparison of data across the two groups indicated that the intervention group showed a better percent reduction in systolic and diastolic BP (−2.7, −3.1) and HbA1c (−1.9) levels as compared to control group (−2.1, −2.4, and −1.4), although the results were not statistically significant [Table 3].
|Table 3: Comparison of percent change in anthropometric and biochemical parameters across the groups|
Click here to view
Among those with diabetes in the intervention group, the mean systolic and diastolic BP reduced from 130.1 ± 15 to 125.3 ± 13.9 (P = 0.007) and 83.3 ± 11.8 to 78.9 ± 8.8 (P = 0.003), respectively. There was no significant change in HbA1c [Table 2].
In the control group, the mean systolic BP reduced from 129.1 ± 16.8 to 124.2 ± 13.3 (P = 0.02) and the mean HbA1c reduced from 7.9 ± 1.6 to 7.7 ± 1.5 (P = 0.03) [Table 2].
Comparison of data across the study groups indicated that the intervention group showed a better percent reduction in systolic and diastolic BP (−3.3, −4.3) compared to control group (−3.1, −2.5). Females with diabetes from the intervention group showed a significant reduction in waist circumference (−3.60) as compared to control participants (7.60) of same category. However, the control group showed a statistically significant percent reduction in HbA1c levels (−2.6; P < 0.05) as compared to the intervention group (3.7) [Table 3].
Smoking behavior and awareness regarding diabetes
Normal glucose tolerance
The number of nonsmokers in the intervention group, increased by 7% compared to only a 2% increase in the control group [Table 4]. Of the participants, 56% in the intervention group (P < 0.0001) and 35% in the control (P = 0.01) group quit smoking compared to baseline. Although a better trend to quit smoking was seen in intervention group as compared to control group, the results were not statistically significant between groups.
|Table 4: Overall percentage of nonsmokers, smokers, and percentage change in awareness regarding diabetes across categories during the study|
Click here to view
The intervention group participants had improved awareness about causes of diabetes and its complications compared to the control group.
A significant number of participants from both intervention (39%; P = 0.04) and control (50%; P = 0.002) groups reported that they had quit smoking postintervention, but the differences between groups were not significant.
Intervention group participants reported significant improvement in awareness regarding most common causes of uncontrolled diabetes (26% change) as compared to control group (1.2% change). On the other hand, control group showed improved awareness regarding causes of diabetes (10.1% change) as compared to intervention group (1.1% change).
Among intervention group, 50% (P = 0.02) reported that they had quit smoking, whereas in control group, only 14% participants reported they had quit smoking.
PA pattern across glycemic categories
Normal glucose tolerance
The intervention group reported 8% change in participants who tend to take stairs instead of using lifts or escalator (P = 0.001) as compared to control group (1.09% change) [Table 5]. The control group, however, reported significant percent change in the participants who exercised regularly during a week (33.3% change; P = 0.002), prefer to walk down small distances for daily chores (21.7% change; P = 0.02), tend to take short walking breaks when working in office or home (26.7% change; P < 0.0001), and helped with household work (18.7% change; P < 0.0001) as compared to intervention group (18.5%, 9.7%, −0.9%, and −8.04% changes, respectively).
|Table 5: Percentage change in physical activity pattern across glycemic categories during the study|
Click here to view
The intervention group reported 48.7% change in participants who take short walking breaks when working in office/home (P < 0.0001). On the other hand, the control group reported 65.4% change in participants who exercised regularly during a week (P < 0.0001).
The intervention group reported a significant percent change (59.1%) in the participants who reported that they exercised as compared to control group (22.7%). The intervention group also reported a 29.2% change in the participants who helped in household chores (P = 0.02) as compared to control group (9.5% change).
Change in eating habits across glycemic categories
Normal glucose tolerance
Intervention group participants reported an improved consumption of daily vegetables (6.1% change) and healthy foods (20.8% change), and a decrease in the consumption of unhealthy foods (−4.3% change) [Table 6]. On the other hand, control group participants reported an improved consumption of fruits (3.9% change), vegetables (3.2% change), and healthy foods (15.1% change), and a marked decrease in the consumption of unhealthy foods (−7.7% change).
|Table 6: Percentage change in eating habits across glycemic categories during the study|
Click here to view
Intervention group participants reported a 23.9% change in the participants consuming one or more fruits daily (P < 0.0001) and decrease in the use of additional salt (−12.5% change, P = 0.02) as compared to control group (3.9% change, −5.4% change) postintervention.
Intervention group participants reported an improved consumption of daily vegetables (14.5% change), healthy foods (25.1% change), and unhealthy foods (−0.8% change). on the other hand, control group participants reported 11.9% change in consumption of vegetables, 13.1% in healthy foods, and marginal decrease in consumption of unhealthy foods (−1.4% change).
Intervention group participants reported 21.4% change in the participants consuming one or more fruits daily (P = 0.03) and significant decrease in the use of additional salt (−38.9% change, P < 0.001) as compared to control group (9.1% change, 31.2% change, respectively) postintervention.
Intervention group participants reported improved consumption of vegetables per day (21.2% change) and other eating habits except for the use of additional salt. On the other hand, control group participants reported improved consumption of fruits (28% change), vegetables (6.1% change), and healthy foods.
Intervention group participants reported a decrease in the consumption of unhealthy foods on daily basis (−11.5% change, P = 0.03) as compared to control group. On the other hand, control group participants reported a decrease in the use of additional salt (−30.0% change, P = 0.02).
Behavior scores across glycemic categories
Moderate-to-vigorous physical activity
Both intervention and control group reported an increase in the number of participants engaged in moderate-to-vigorous physical activity (MVPA) [Table 7]. Prediabetes category of intervention group showed a significant increase in MVPA compared to control group, highlighting a better adoption of lifestyle habits.
|Table 7: Change in moderate-to-vigorous physical activity, mean eating score, and mean behavior score across glycemic categories during the study|
Click here to view
The mean eating score increased significantly for all the glycemic categories across both the study groups postintervention. The mean score was higher in prediabetes and DM participants as compared to NGT participants across both study groups.
Total behavior score
The mean total behavior score increased significantly across glycemic categories in both study groups. The mean score was higher in prediabetes and DM participants as compared to NGT participants in both the groups.
| Discussion|| |
The salient findings of the study are as follows:
Similar number of participants met the study goals across the groups, thus making it a null effect study.
However, the study results indicated that across glycemic categories, prediabetes and DM participants were more inclined toward making positive lifestyle changes as compared to participants in NGT category.
The myArogya app study included majorly young population aged between 20 and 35 years and nearly half of them (57%) were at the risk of diabetes (IDRS 30–59). This can be attributed to the changing lifestyle and eating habits of the participants who primarily belonged to the information technology (IT) sector. The “myArogya” effectiveness study gauged the impact of a lifestyle based smartphone app without a human interface on the anthropometric, biochemical, dietary, PA, and smoking behaviors of the participants in a RCT across categories of glycemia.
The intervention group participants belonging to the DM category showed a better reduction in weight, waist circumference, and body mass index (BMI) whereas in the control group there was a reduction in weight and BMI. On the other hand, weight, waist circumference, and BMI increased in NGT category of both the study groups postintervention. Kodama et al. did a meta-analysis of 23 studies including 8697 participants to study the effect of web-based lifestyle modification on weight control. They concluded that the internet component in obesity treatment programs has a modest effect on weight control. However, the effect was inconsistent, largely depending on the type of usage of the internet or the period of its use.
The intervention group showed a better reduction in both systolic and diastolic BP across the glycemic categories, compared to the control group. However, the control group showed more prominent reductions in BP compared to intervention that may be ascribed to a positive decline in the use of additional salt. A longitudinal study that evaluated the effect of the daily use of a mobile phone-based self-management support system for hypertension showed the daily use of the support system significantly reduced BP between baseline and week 8.
Postintervention, the HbA1c levels increased in the NGT category but stayed within the normal range across both study groups. Although there was a significant reduction in HbA1c levels in prediabetes category across both groups, among DM category HbA1c levels reduced significantly in control group while increased slightly in the intervention group. As HbA1c was one of the primary outcomes of our study to prove app effectiveness, our hypothesis was disproved. Previous studies have shown similar results. Holmen et al. evaluated the effectiveness of a mobile-phone-based self-management system on HbA1c levels of 150 participants who were divided into three groups: intervention group who received mobile app, intervention group who received mobile app plus health counseling, and control group. Their results showed no difference in HbA1c levels between the groups, but there was an increase in the self-management domain of the skill of intervention group who received mobile app plus health counseling, which was very similar to what was seen in our study.
Our study showed a decline in blood cholesterol levels across the glycemic categories in both study groups postintervention. The most significant lifestyle behavior change was that overall, both the groups showed a positive drift in decreasing smoking behavior. Knowing the importance of tobacco cessation, 13 tobacco cessation clinics were started in 2002 by the Ministry of Health and Family Welfare, Government of India. Research data from various studies have shown that behavioral interventions can help in smoking cessation. Our study reiterated the same.
The participants from both the groups showed a positive inclination toward some or the other kind of PA. Similar results on PA behaviors have been reported in various other studies.,, In an RCT conducted in Auckland, New Zealand, by Direito et al., participants aged 14–17 years were divided into three groups: (1) use of an immersive app (Zombies, Run), use of a nonimmersive app (Get Running), or usual behavior (control). The primary outcome was cardiorespiratory fitness and secondary outcomes were PA levels. The results showed that although apps have the ability to maximize reach at a low cost, the rational approach of the researcher using readily available commercial apps as a stand-alone instrument did not have a significant effect on fitness. However, participants did show an increased interest in the future to use PA apps, emphasizing a possibly important role of these tools in a multidimensional approach to increase fitness, promote PA, and consequently reduce the adverse health outcomes associated with the insufficient activity.
The intervention group participants reported a significant increase in the consumption of one or more fruits daily and decrease in the use of additional salt in NGT and prediabetes category. Similarly, control participants reported a significant decrease in the use of additional salt in DM category. Both the groups reported a decline in consumption of unhealthy food across the glycemic categories. Overall participants from both the groups showed a positive trend toward healthy eating habits with intervention group participants doing better than the control group.
A study evaluated the relation between diet and nutrition-related mobile apps and behavior change in 217 participants who answered questions on demographics, use of diet and nutrition apps in the past 6 months, engagement and likability of apps, and changes in the participant’s dietary behaviors. The results showed that most of the study participants agreed or strongly agreed with statements regarding app use increasing their motivation to eat a healthy diet, improving their self-efficacy, and increasing their desire to set and achieve healthy diet goals. In addition, the majority of participants strongly agreed that using diet/nutrition apps led to changes in their behaviour., Our work in mHealth previously as well as in this study are also consistent with these results.
It was observed that approximately the same percentage of control group (48%) and intervention group (47%) participants achieved two or more goals during the study thus making this a null effect study.
Among lifestyle goals, both intervention and control group participants reported an increase in the number of participants engaged in MVPA, mean eating score, and mean total behavior score. This is in accordance to most studies that report that though mobile health (mHealth) possibly will not be able to replace professional health care, it is empowering individuals with information related for the prevention and management of various ailments and is enabling them to make wise decisions for positive health.,,,,,,,
The strength of this study was the large sample size at baseline and randomization of participants with the risk of DM in intervention and control group, together with the objective measurement of anthropometric and biochemical parameters. Pretested text messages in four different modules augmented the approach of the app to all noncommunicable diseases and not just diabetes.
As this was not a cluster randomized trial, control group participants also had access to the app from their friends/colleagues within the worksite who were in the intervention group. This directs toward a possible cross-contamination as this was a worksite-based study. Our results also indicate a stronger inclination among the control group participants to make positive lifestyle changes.
However, the high dropout percentage due to various reasons was our biggest limitation. A few reasons have been listed here. Some study sites showed unwillingness to continue with participation postintervention. Most of the young-aged study participants working in the IT sector believed that there was nothing wrong with their health. Also, they keep experimenting with different apps rather than sticking to just one. We also lost a few of study participants owing to high job attrition rates in IT industry. Not all workplaces were equipped with Wi-Fi or internet access, restricting usage of the app. Some of the larger companies where the dropout percent was very heavy did attribute this to the fact that the employees were inundated with free health programs. Also, our app lacked personalized touch. Personalization seems to be the key element in engaging a diverse group of participants. This is supported by Tang et al., who found that weight loss app users appreciated simplicity, personalized features, and accessibility of the app. In another article by Muralidharan S et al., it was concluded that along with mHealth technology, support by a health care professional results in added positive outcomes inpatients.
| Conclusions|| |
“myArogya” effectiveness study motivated both intervention and control group participants to adopt a healthy lifestyle. Mean behavior scores indicate that both intervention and control group participants were equally driven to adopt a healthier lifestyle. In conclusion, the myArogya app showed a significant improvement in primary outcomes in the prediabetes category of intervention group and marginal improvements in most parameters across other glycemic categories in both study groups. Overall, even though it was null-effect study, the app has empowered individuals with information related to prevention and management of NCDs and this prompted users to make positive lifestyle changes.
We are grateful to Cigna Foundation for supporting projects that advance global health. We also wish to record the support of Nandini Ganesh from Arogya World for this project and her untiring efforts to co-ordinate all site visits.
Financial support and sponsorship
The myArogya study was supported by a grant from Cigna Foundation to Arogya World.
Conflicts of interest
There are no conflicts of interest.
| References|| |
International Diabetes Federation. Diabetes Atlas 7th edition 2015. IDF ;2015. Available from: http://www.diabetesatlas.org/. [Last accessed on 2019 Jan 29].
Kaveeshwar SA, Cornwall J. The current state of diabetes mellitus in India. The Australas Med J 2014;7:45-8.
Anjana RM, Pradeepa R, Deepa M, Datta M, Sudha V, Unnikrishnan R, et al
; ICMR–INDIAB Collaborative Study Group. Prevalence of diabetes and prediabetes (impaired fasting glucose and/or impaired glucose tolerance) in urban and rural India: Phase I results of the Indian Council of Medical Research-INdia DIABetes (ICMR-INDIAB) study. Diabetologia 2011;54:3022-7.
Ramachandran A, Snehalatha C, Vijay V. Low risk threshold for acquired diabetogenic factors in Asian Indians. Diabetes Res Clin Pract 2004;65:189-95.
Anjana RM, Deepa M, Pradeepa R, Mahanta J, Narain K, Das HK, et al
; ICMR–INDIAB Collaborative Study Group. Prevalence of diabetes and prediabetes in 15 states of India: Results from the ICMR-INDIAB population-based cross-sectional. Lancet Diabetes Endocrinol 2017;5:585-96.
Powers MA, Bardsley J, Cypress M, Duker P, Funnell MM, Fischl AH, et al
. Diabetes Self-management Education and Support in Type 2 Diabetes: A joint position statement of the American Diabetes Association, the American Association of Diabetes Educators, and the Academy of Nutrition and Dietetics. Diabetes Care 2015;38:1372-82.
Haas L, Maryniuk M, Beck J, Cox CE, Duker P, Edwards L, et al
. National standards for diabetes self-management education and support. Diabetes Care 2013;36:S100-8.
Weber MB, Ranjani H, Staimez LR, Anjana RM, Ali MK, Narayan KM, et al
. The stepwise approach to diabetes prevention: Results from the D-CLIP randomized controlled trial. Diabetes Care 2016;39:1760-7.
Ramachandran A, Snehalatha C, Ram J, Selvam S, Simon M, Nanditha A, et al
. Effectiveness of mobile phone messaging in prevention of type 2 diabetes by lifestyle modification in men in India: A prospective, parallel-group, randomised controlled trial. Lancet Diabetes Endocrinol 2013;1:191-8.
Ramachandran A, Snehalatha C, Mary S, Mukesh B, Bhaskar AD, Vijay V; Indian Diabetes Prevention Programme (IDPP). The Indian Diabetes Prevention Programme shows that lifestyle modification and metformin prevent type 2 diabetes in Asian Indian subjects with impaired glucose tolerance (IDPP-1). Diabetologia 2006;49:289-97.
Gillies CL, Abrams KR, Lambert PC, Cooper NJ, Sutton AJ, Hsu RT, et al
. Pharmacological and lifestyle interventions to prevent or delay type 2 diabetes in people with impaired glucose tolerance: Systematic review and meta-analysis. BMJ 2007;334:299.
Kirwan M, Vandelanotte C, Fenning A, Duncan MJ. Diabetes self-management smartphone application for adults with type 1 diabetes: Randomized controlled trial. J Med Internet Res 2013;15:e235.
Free C, Phillips G, Watson L, Galli L, Felix L, Edwards P, et al
. The effectiveness of mobile-health technologies to improve health care service delivery processes: A systematic review and meta-analysis. PLoS Med 2013;10:e1001363.
Patient Adoption of mHealth: Use, Evidence and Remaining Barriers to Mainstream Acceptance. Available from: https://www.iqvia.com/-/media/iqvia/pdfs/institute-reports/patient-adoptionof-mhealth.pdf?la=en&hash=B3ACFA8ADDB143F29EAC0C33- D533BC5D7AABD689. Last accessed on 2019 January.
Pfammatter A, Spring B, Saligram N, Davé R, Gowda A, Blais L, et al
. mHealth intervention to improve diabetes risk behaviors in India: A prospective, parallel group cohort study. J Med Internet Res 2016;18:e207.
Mohan V, Deepa R, Deepa M, Somannavar S, Datta M. A simplified Indian diabetes risk score for screening for undiagnosed diabetic subjects. J Assoc Physicians India 2005;53:759-63.
Quest Diagnostics: Interpretive Guide. Available from: https://www.questdiagnostics.com/hcp/intguide/hcp_ig_main2.html. Last accessed on 2019 January.
Kodama S, Saito K, Tanaka S, Horikawa C, Fujiwara K, Hirasawa R, et al
. Effect of web-based lifestyle modification on weight control: A meta-analysis. Int J Obesity 2012;36:675-85.
Bengtsson U, Kjellgren K, Hallberg I, Lindwall M, Taft C. Improved blood pressure control using an interactive mobile phone support system. J Clin Hypertens (Greenwich) 2016;18:101-8.
Holmen H, Torbjørnsen A, Wahl AK, Jenum AK, Småstuen MC, Arsand E, et al
. A mobile health intervention for self-management and lifestyle change for persons with type 2 diabetes, part 2: One-year results from the Norwegian Randomized Controlled Trial RENEWING HEALTH. JMIR Mhealth Uhealth 2014;2:e57.
Thankappan KR. Tobacco cessation in India: A priority health intervention. Indian J Med Res 2014;139:484-6.
] [Full text]
Luis G, Jose I, Cristina A, María C, Jose-Angel M, Irene R, et al
. Long-term effectiveness of a smartphone app for improving healthy lifestyles in general population in primary care: Randomized controlled trial (Evident II Study). JMIR Mhealth Uhealth 2018;6:e107.
Seifert A, Schlomann A, Rietz C, Schelling HR. The use of mobile devices for physical activity tracking in older adults’ everyday life. Digit Health 2017;3:2055207617740088.
Direito A, Jiang Y, Whittaker R, Maddison R. Apps for improving fitness and increasing physical activity among young people: The AIMFIT pragmatic randomized controlled trial. J Med Internet Res 2015;17:e210.
Sarah M, Thomas NR, Maya M, Sarah F, Stephen S, Christopher DG. Effect of a mobile app intervention on vegetable consumption in overweight adults: A randomized controlled trial. Int J Behav Nutr Phys Act 2017;14:125.
West JH, Belvedere LM, Andreasen R, Frandsen C, Hall PC, Crookston BT. Controlling your “App”etite: How diet and nutrition-related mobile apps lead to behavior change. JMIR MHealth and Uhealth 2017;5:e95.
Tang J, Abraham C, Stamp E, Greaves C. How can weight-loss app designers’ best engage and support users? A qualitative investigation. Br J Health Psychol 2015;20:151-71.
Muralidharan S, Ranjani H, Anjana RM, Allender S, Mohan V. Mobile health technology in the prevention and management of type 2 diabetes. Indian J Endocrinol Metab 2017;21:334-40.
[Figure 1], [Figure 2]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7]