A randomized controlled trial tested a virtual assistance-based lifestyle intervention to reduce cardiometabolic risk factors in young IT employees in India. Employees with ≥3 risk factors were randomized to a control or intervention group. The intervention group received lifestyle advice plus mobile/email messages for 1 year, while the control received only initial advice. The primary outcome was change in overweight-obesity prevalence. After 1 year, overweight-obesity decreased in the intervention group but increased in controls, with an 11.2% difference between groups. The intervention was effective, cost-effective, and acceptable for reducing risk factors in young IT employees.

1. Elsevier Editorial System(tm) for The Lancet
Manuscript Draft
Manuscript Number:
Title: A virtual assistance based lifestyle intervention is effective in reducing cardiometabolic risk
factors in young employees of Information Technology industry in India (LIMIT): a pragmatic
randomised controlled trial
Article Type: Article (Randomised Controlled Trial)
Corresponding Author: Dr. Chittaranjan S Yajnik, MD, FRCP
Corresponding Author's Institution: King Edward Memorial Hospital and Research Centre
First Author: Tejas Y Limaye, MSc, RD
Order of Authors: Tejas Y Limaye, MSc, RD; Kalyanaraman Kumaran, DM; Charudatta V Joglekar, MS;
Dattatray S Bhat, MSc; Ravindra L Kulkarni, MD; Arun S Nanivadekar, MD; Chittaranjan S Yajnik, MD,
FRCP
Abstract: Background: We investigated a virtual assistance based lifestyle intervention to reduce
cardiometabolic risk factors in young high-risk employees of Information Technology (IT) industry in
India.
Methods: LIMIT (LIfestyle Modification in IT) was a parallel-group, randomised controlled trial
conducted between May 2012 and October 2013 (CTRI registration: CTRI/2015/01/005376). IT
employees from two industries with ≥3 cardiometabolic risk factors (impaired fasting glucose, high
blood pressure, hypertriglyceridemia, high LDL cholesterol, low HDL cholesterol, overweight-obesity,
and family history of cardiometabolic disease) were randomised into control or intervention groups
(1:1) using computer-generated sequence. After initial lifestyle advice, the intervention group
additionally received reinforcement through mobile phone messages (three/week) and e-mails
(two/week) for one year. Field staff and participants could not be masked to group allocation;
laboratory technicians and statisticians were masked. The primary outcome was change in prevalence
of overweight-obesity, analysed by intention to treat.
Findings: Of 437 employees screened (mean age 36*2±9*3 years; 74*8% men), 265 (61*0%) were
eligible and randomised into control (n=132) or intervention (n=133) groups. The groups were
comparable at baseline. After one year, the number of overweight-obese participants decreased from
104 (78*2%) to 96 (72*2%) in the intervention group (p=0*021) while it increased from 101 (76*5%)
to 110 (83*3%) in the control group (p=0*004); risk difference 11*2% (95% CI 1*2-21*1; p=0*042).
The number needed to treat (NNT) to reverse one case of overweight-obesity in one year was 9 (95%
CI 5-82). The incremental cost per NNT was INR 10665 ($171). There were no adverse events and
98*0% participants opted for continuation of the virtual assistance.
Interpretation: Lifestyle intervention using virtual assistance is effective, cost-effective, and acceptable
in reducing cardiometabolic risk factors in young IT employees, and has potential for scaling up.
Funding: Diabetes Unit, KEM Hospital Research Centre, Pune; Department of Science and Technology,
New Delhi.

3. 1
A virtual assistance based lifestyle intervention is effective in reducing cardiometabolic risk factors
in young employees of Information Technology industry in India (LIMIT): a pragmatic
randomised controlled trial
Tejas Limaye (MSc)1
, Kalyanaraman Kumaran (DM)1,2
, Charudatta Joglekar (MS)1
, Dattatray Bhat
(MSc)1
, Ravindra Kulkarni (MD)3
, Arun Nanivadekar (MD)4
, Prof. Chittaranjan Yajnik (MD)1
1
Diabetes unit, King Edward Memorial (KEM) Hospital Research Centre Pune, India
2
MRC Lifecourse Epidemiology Unit, University of Southampton, UK
3
Just for Hearts Healthcare Pvt. Ltd., Pune, India
4
Medical research consultant, Mumbai, India
Correspondence to:
Prof. Chittaranjan Yajnik
Diabetes Unit, 6th
floor, Banoo Coyaji Building,
KEM Hospital Research Centre,
Rasta Peth, Pune – 411011
Maharashtra, India
E-mail: csyajnik@hotmail.com
Telephone: +91-20-26111958; Fax: +91-20-26111958.
Manuscript including tables and figures

4. 2
Summary
Background: We investigated a virtual assistance based lifestyle intervention to reduce cardiometabolic
risk factors in young high-risk employees of Information Technology (IT) industry in India.
Methods: LIMIT (LIfestyle Modification in IT) was a parallel-group, randomised controlled trial
conducted between May 2012 and October 2013 (CTRI registration: CTRI/2015/01/005376). IT
employees from two industries with ≥3 cardiometabolic risk factors (impaired fasting glucose, high blood
pressure, hypertriglyceridemia, high LDL cholesterol, low HDL cholesterol, overweight-obesity, and
family history of cardiometabolic disease) were randomised into control or intervention groups (1:1)
using computer-generated sequence. After initial lifestyle advice, the intervention group additionally
received reinforcement through mobile phone messages (three/week) and e-mails (two/week) for one
year. Field staff and participants could not be masked to group allocation; laboratory technicians and
statisticians were masked. The primary outcome was change in prevalence of overweight-obesity,
analysed by intention to treat.
Findings: Of 437 employees screened (mean age 36·2±9·3 years; 74·8% men), 265 (61·0%) were
eligible and randomised into control (n=132) or intervention (n=133) groups. The groups were
comparable at baseline. After one year, the number of overweight-obese participants decreased from 104
(78·2%) to 96 (72·2%) in the intervention group (p=0·021) while it increased from 101 (76·5%) to 110
(83·3%) in the control group (p=0·004); risk difference 11·2% (95% CI 1·2–21·1; p=0·042). The number
needed to treat (NNT) to reverse one case of overweight-obesity in one year was 9 (95% CI 5–82). The
incremental cost per NNT was INR 10665 ($171). There were no adverse events and 98·0% participants
opted for continuation of the virtual assistance.
Interpretation: Lifestyle intervention using virtual assistance is effective, cost-effective, and acceptable
in reducing cardiometabolic risk factors in young IT employees, and has potential for scaling up.
Funding: Diabetes Unit, KEM Hospital Research Centre, Pune; Department of Science and Technology,
New Delhi
Keywords: Information technology industry, lifestyle modification, overweight-obesity, virtual
assistance, technology based intervention, pragmatic intervention

5. 3
Introduction
The rising incidence of obesity, type 2 diabetes (T2D) and other cardiometabolic diseases in India has
partly been attributed to rapid socio-economic transition, adoption of increasingly calorie dense diets and
physical inactivity.1
Young and economically productive populations are particularly affected, resulting in
significant socio-economic implications for individuals and society.2
Targeting this population with cost-
effective primary prevention strategies may blunt this impact.
Several clinical trials have shown that lifestyle modification can reduce conversion from prediabetes to
T2D by almost 50%.3–5
They demonstrated that the main drivers of diabetes prevention are weight loss
and physical activity. However, these trials were expensive and labour intensive on account of
individualised counseling, supervised exercise sessions, and extensive personal follow-up.6,7
Translating
these trials to real world settings remains a challenge. Thus novel and effective approaches which are
practical, affordable, and scalable need to be developed and tested.
In the last two decades, the Information Technology (IT) sector in India has expanded substantially and
has experienced considerable economic growth.8
Work environment in the IT sector exposes the
employees to a variety of risk factors for T2D including sedentary lifestyle, long working hours, erratic
eating habits, and high stress levels.9
On this background, we tested the effectiveness of a virtual
assistance based lifestyle intervention to reduce cardiometabolic risk factors in young, high-risk IT
employees.
Methods
Study design and participants ‘LIMIT’ was a parallel-group, randomised controlled trial conducted
between May 10, 2012, and October 20, 2013. We approached 10 IT industries in Pune (India) of which 2
agreed to participate in the trial. The trial started with a weeklong teaser campaign to sensitise the
employees, followed by voluntary registrations. A trained team (a research fellow, 2 research assistants,
and 2 laboratory technicians) visited the worksites and screened the participants using standardised
anthropometric measurements, blood pressure (BP), fasting blood investigations, and a structured
questionnaire (supplementary material 1).
Those with known diabetes or hypertension, those who were newly diagnosed with T2D (fasting plasma
glucose ≥7 mmol/L)10
or stage-2 hypertension (systolic BP ≥160 and/or diastolic BP ≥100 mmHg),11
and
those on lipid lowering drugs were excluded. Newly diagnosed participants were referred to their family
physicians for treatment. Other exclusion criteria included pregnancy, triglycerides ≥5·7 mmol/L,12
severe
anemia (hemoglobin <80 g/L),13
major illness, and disability which can restrict physical activity.

6. 4
Participants with less than three cardiometabolic risk factors (listed in table 1) were also excluded.
Participants who had ≥3 cardiometaboilc risk factors (table 1) were eligible for randomisation. The study
protocol was approved by the Ethics Committee of KEM Hospital Research Centre, Pune; and informed
written consent was obtained from all participants.
Table 1: List of cardiometabolic risk factors used in inclusion criteria
No. Risk factor Cutoffs
1 Family history 1st
degree family history of T2D and/or cardiovascular disease
2 Overweight-obesity BMI ≥25 (kg/m2
)14
3 Central obesity waist circumference ≥90·0 (cm) in men or ≥80·0 (cm) in women15
4 IFG FPG 5·6 – 6·9 (mmol/L)10
5 Raised BP systolic BP ≥130 and/or diastolic BP ≥85 (mmHg)11,15
6 High triglycerides plasma triglycerides ≥1·7 (mmol/L)15
7 Low HDL cholesterol HDL cholesterol <1·0 (mmol/L) in men or <1·3 (mmol/L) in women15
8 High LDL cholesterol LDL cholesterol ≥3·4 (mmol/L)12
BMI: Body Mass Index, BP: Blood Pressure, FPG: Fasting Plasma Glucose, IFG: Impaired Fasting
Glucose, T2D: Type 2 Diabetes
Randomisation and masking A research assistant not involved in data analysis allocated the eligible
participants to intervention or control groups (1:1) using a centrally generated computer randomisation
scheme (randomly permuted blocks, block size=8; www.randomization.com).16
Field staff and
participants could not be masked to group allocation due to the nature of the intervention but the
laboratory staff and statisticians were masked until the end of analysis. The participants were asked not to
disclose their allocation group to the laboratory personnel during sample collection and were identified
with a study number during laboratory analysis. The two groups were identified numerically during
statistical analysis without revealing group allocation.
Procedures Height was measured (nearest 0·1 cm) using a wall-mounted stadiometer (CMS instruments,
London, UK) and weight (nearest 0·1 kg) by a digital scale (Omron HBF 375). Waist circumference was
measured (nearest 0·1 cm) at a point midway between the lower costal margin and the superior iliac crest
at the end of a normal expiration using a non-stretchable measuring tape (CMS Instruments, London,
UK). BP was measured in the sitting position, on the right arm, after 5 minutes rest using a digital
monitor (UA 767PC, A and D Instruments Ltd, Abingdon, Oxford, UK). The average of two readings
taken 5 minutes apart was used for analysis. Fasting blood was collected in 10 ml BD vacutainer tubes
(EDTA) and transported to our laboratory in an icebox. Hemoglobin was measured on a Beckman Coulter
Analyser (AC.T diff; Miami, Florida). The samples were then centrifuged at 3000g at 4°C for 10 minutes.

7. 5
The plasma levels of glucose, triglycerides, total cholesterol, HDL cholesterol, and creatinine were
measured using standard enzymatic methods on autoanalyser (Hitachi 902, Hitachi Corporation, Japan)
with a coefficient of variation <4% for all. LDL was calculated by Friedewald formula.17
A pretested
questionnaire was used to record demographic details, medical and family history, lifestyle information
(diet, physical activity and substance use), and awareness of diabetes. An awareness score was calculated
based on the number of correct responses to questions regarding implications of diabetes.18
At baseline, all participants received advice on lifestyle modification in group sessions. Overweight-obese
participants were advised to lose weight and achieve a body mass index (BMI) of <25kg/m2
. We also
identified four lifestyle modification goals for all participants: to achieve and/or maintain minimum 150
minutes of moderate physical activity per week, to increase consumption of fiber rich foods to ≥8
servings per week, to reduce consumption of calorie dense foods to ≤4 servings per week, and achieve
awareness score ≥75%. These goals were based on baseline observations and/or standard guidelines.19
Written information on diet and physical activity was distributed at the end of the session.
In addition, participants in the intervention group received regular reminders on lifestyle modification
through mobile phone messages and e-mails for one year. Following a survey of participants’ preferences,
three mobile phone messages and two e-mails were sent per week between 10.00–13.00 h. Smokers
received an additional message every weekend. Mobile phone messages contained a maximum of 160
characters. E-mails contained info-graphics (visual representation of the information). E-mails were sent
to both personal and official e-mail addresses of participants. Participants did not receive same message
or e-mail during the entire intervention. Additional support through a website and a Facebook page was
also offered to the participants in the intervention group. Messages and e-mails on healthy lifestyle, diet,
physical activity, stress management and weight loss were sent in rotation interspersed with messages and
e-mails of encouragement (supplementary material 2 and 3). An individualised target body weight was set
for overweight-obese subjects with an aim to lose a minimum of 5% of their baseline weight.3
The
prescribed lifestyle changes were in line with those used in previous diabetes prevention trials.3–5
Participants’ queries to specific e-mails or mobile phone messages were resolved within 24 hours.
Participants in the intervention group were advised not share the mobile phone messages, e-mails, and the
details of website and Facebook page with their colleagues to prevent contamination. Of a total of 150
mobile phone messages and 100 e-mails sent during the year, one-tenth (15 mobile phone messages and
10 e-mails) requested a reply. Compliance was calculated based on the response to these requests.
Participants with ≥75% compliance were considered ‘compliers’.

8. 6
We reassessed all the participants every three months. Anthropometry and BP measurements were
conducted at all follow-up visits. Biochemistry, lifestyle information, diabetes awareness and
acceptability of the intervention were measured at one year.
Costs of recruitment, intervention, and follow-up were calculated considering the direct medical costs
incurred over the one-year study period (including research costs). Indirect and societal costs were not
considered. As the primary outcome was prevalence of overweight-obesity, the cost-effectiveness of the
intervention was calculated as the incremental cost to reverse one case of overweight-obesity within the
one-year trial period (difference in costs between intervention and control groups multiplied by the
number needed to treat). Costs are expressed in Indian rupees (INR) or the 2015 U.S. dollar equivalent.
Outcomes The primary outcome of interest was prevalence of overweight-obesity (BMI ≥25 kg/m2
).
Secondary outcomes included change in weight, waist circumference, BP, glucose, lipids, lifestyle
choices (physical activity, frequencies of calorie dense and fiber rich foods, smoking), diabetes awareness
score, and acceptability of the intervention. We recorded adverse events and treatment of intercurrent
illnesses, if any, at each follow-up visit.
Statistical analysis Pre-trial analysis of annual health records of IT employees from two different IT
industries showed a prevalence of 75% overweight-obesity in high-risk employees. We set out to detect a
difference of 25% in prevalence of overweight-obesity between the intervention and control groups (1:1)
at 1 year at 5% significance and 80% power. We assumed a dropout of 25% due to high attrition rates and
frequent travel schedules in IT professionals.20
This gave us a required sample size of 132 individuals in
each group. Analysis was done by intention to treat. The participants who were lost to follow-up were
analysed by ‘last observation carried forward’ method. The baseline values were carried forward for those
who could not attend any follow-up.
The data are presented as mean (SD) for normally distributed variables and as median (25th
–75th
percentile) for skewed variables. Skewed variables (triglycerides and HDL cholesterol) were log
transformed to ensure normality during statistical analysis. Baseline measurements were compared using
independent sample t-test for continuous variables, and chi-square test for categorical variables.
Comparisons between baseline and subsequent three month measurements were made using paired t-test.
The significance of the difference between means of the two groups was assessed by ANOVA with
adjustments for confounding variables (age and gender) as appropriate. McNemar test was used to
compare paired proportions. The number needed to treat (NNT) with 95% CI was calculated as the
inverse of the absolute risk reduction and its 95% CI.

9. 7
Analysis was carried out with the statistical package for social sciences for windows (SPSS, Chicago III),
version 16. The trial was registered with the Clinical Trial Registry of India (ctri.nic.in), number
CTRI/2015/01/005376.
Role of the funding source The funder of the study had no role in study design, data collection, data
analysis, data interpretation, or writing of the report. The corresponding author had full access to all the
data in the study and had final responsibility for the decision to submit for publication.
Results
Figure 1 shows the trial profile; 437 participants (mean age 36·2±9·3 years; 74·8% men) were screened.
Seventy-four (16·9%) participants with diabetes, hypertension, triglycerides ≥5·7 mmol/L or on lipid
lowering drugs, and 98 (22·4%) with less than three risk factors were excluded. Thus 265 (60·6%)
participants (mean age 36·2±8·0 years; 72·5% men) with ≥3 risk factors were eligible and randomly
assigned to intervention (n = 133) or control (n = 132) groups. Of them, 205 (77·4%) had overweight-
obesity, 233 (87·9%) had central obesity, 189 (71·3%) had low HDL cholesterol, 71 (26·8%) had high
LDL cholesterol, 70 (26·4%) had high triglycerides, 67 (25·3%) had raised BP, and only three (1·1%)
participants had impaired fasting glucose (cutoffs are given in table 1). Seventy one (26·8%) participants
had metabolic syndrome.13
There were no significant differences in baseline characteristics between the
two groups (table 2).
A total of 203 (76·6%) participants [intervention group: 105 (78.9%), control group: 98 (74·2%);
p=0·366] completed the trial (figure 1). Job changes, travel, and busy work schedules were the most
common reasons for missing follow-up assessments. Those who were lost to follow-up were no different
than who continued in the trial in terms of age, gender, years of education, weight, BMI, waist
circumference, BP, FPG, triglycerides, total, HDL and LDL cholesterol levels at baseline. The final
(intention to treat) analysis includes all 265 randomised participants.

10. 8
Figure 1: Trial Profile
At three months the difference in prevalence of overweight-obesity from baseline was not significant in
either group, at six months the difference was significant only in the intervention group while at nine
months the difference was significant in both the groups (figure 2). After one year, the number of
overweight-obese participants decreased from 104 (78·2%) to 96 (72·2%) in the intervention group
(p=0·021) while it increased from 101 (76·5%) to 110 (83·3%) in the control group (p=0·004); risk
difference 11·2% (95% CI 1·2 – 21·1; p=0·042). The NNT to reverse one case of overweight-obesity in
one year was 9 (95% CI 5 – 82).

11. 9
Figure 2: Change in prevalence of overweight-obesity (BMI ≥25 kg/m2
) at different timepoints
At one year, 15 (14·4%) overweight-obese participants in the intervention group lost ≥5% of their
baseline weight while 21 (13·5%) lost between 2·5 to 5% of baseline weight. None of the overweight-
obese participants in the control group could achieve ≥5% weight loss; 9 (8·9%) lost 2·5 to 5% of their
baseline weight.
At six months, the intervention group had significantly greater reductions in weight [-1·1 (2·4) vs +0·5
(2·1) kg, p<0·001], waist circumference [-1·5 (2·6) vs +0·5 (1·5) cm, p<0·001], systolic BP [-1·9 (7·6) vs
+0·7 (9·3) mmHg, p=0·012], and diastolic BP [-1·3 (5·7) vs +0·4 (6·7) mmHg, p=0·033] compared to
controls. Improvements were sustained at one year with exception of systolic and diastolic blood pressure
(table 2). Proportion of participants with metabolic syndrome remained the same in both groups.

12. 10
Table 2: Comparison between intervention and control group
Intervention group (n= 133, 74·4% men) Control group (n= 132, 70·5% men) P (for
difference in
∆ between
groups)
Baseline 1-Y ∆ (95% CI) p Baseline 1-Y ∆ (95% CI) p
Demographics
Age (Y) 36·8 (7·2) 35·7 (8·1) ---
Family history 63 (47·4)* 68 (51·5)* ---
Anthropometry
Weight (kg) 74·2 (10·1) 73·2 (10·2) -1·0 (-1·5, -0·5) <0·001 75·4 (12·8) 76·1 (13·3) 0·7 (0·3, 1·2) 0·001 <0·001
BMI (kg/m2
) 27·0 (3·2) 26·6 (3·2) -0·4 (-0·6, -0·2) <0·001 27·3(3·5) 27·6 (3·7) 0·3 (0·1, 0·4) 0·001 <0·001
BMI ≥25 (kg/m2
) 104 (78·2)* 96 (72·2)* -6·0 (-11·3, -0·7) 0·004 101 (76·5)* 110 (83·3)* 6·8 (1·7, 11·8) 0·021 0·020
Waist circumference (cm) 95·6 (7·3) 93·9 (7·5) -1·7 (-2·3, -1·2) <0·001 96·1 (9·3) 96·6 (9·6) 0·5 (0·1, 0·9) 0·013 <0·001
BP
Systolic (mmHg) 113·2 (11·7) 113·1 (13·4) -0·1 (-1·9, 1·8) 0·936 113·3 (11·7) 114·1 (12·7) 0·9 (-0·8, 2·5) 0·293 0·452
Diastolic (mmHg) 77·6 (8·9) 76·2 (9·9) -1·4 (-2·8, -0·1) 0·035 76·9 (8·9) 76·7 (9·9) -0·3 (-1·5, 1·0) 0·682 0·191
Biochemistry
FPG (mmol/L) 4·12 (0·54) 4·31 (0·52) 0·20 (0·12, 0·28) <0·001 4·11 (0·52) 4·44 (0·57) 0·33 (0·25, 0·42) <0·001 0·022
Triglycerides (mmol/L)
1·22
(0·95–1·77)†
1·31
(0·95–1·81)†
0·07 (-0·002, 0·15) 0·057
1·13
(0·84–1·63)†
1·24
(0·88–1·65)†
0·12 (0·02, 0·23) 0·023 0·467
Total cholesterol (mmol/L) 4·65 (0·93) 4·48 (0·88) -0·17 (-0·25, -0·10) <0·001 4·55 (0·78) 4·53 (0·79) -0·03 (-0·10, 0·05) 0·489 0·006
LDL cholesterol (mmol/L) 3·00 (0·81) 2·81 (0·76) -0·19 (-0·26, -0·12) <0·001 2·92 (0·63) 2·85 (0·64) -0·08 (-0·15, -0·01) 0·029 0·023
HDL cholesterol (mmol/L)
0·98
(0·88–1·10)†
0·96
(0·88–1·09)†
-0·02
(-0·03, 0·01)
0·120
0·98
(0·85–1·11)†
0·98
(0·85–1·11)†
-0·01
(-0·02, 0·01)
0·522 0·401
Lifestyle
Exercise (≥150min/week) 30 (22·6) * 57 (42·9) * 20·3 (11·8, 28·7) <0·001 26 (19·7) * 29 (22·0) * 2·3 (-4·2, 8·8) 0·607 <0·001
Intake of fiber rich foods (≥8
servings/week)
19 (14·3)* 32 (24·1)* 9·8 (3·6, 16·0) 0·001 20 (15·2)* 20 (15·2)* 0·0 (-6·7, 6·7) 0·999
0·140
Intake of calorie dense
foods (≤4 servings/week)
18 (13·5)* 39 (29·3)* 17·1 (9·6, 24·5) <0·001 10 (7·6)* 14 (10·6)* 3·0 (-1·9, 7·9) 0·289
0·140
Awareness score (≥75%) 4 (3·0)* 41 (30·8)* 27·8 (19·5, 36·2) <0·001 9 (6·8)* 13 (9·8)* 3·0 (-1·3, 7.4) 0·219 <0·001
Smokers 15 (11·3)* 11 (8·2)* -3·1 (-6·7, 0·6) 0·125 22 (16·8)* 22 (16·8)* 0 (-2·8, 2·8) 0·999 0·600
Numbers are mean (SD), †median (IQR), *N (%), BMI: Body Mass Index, BP: Blood Pressure, FPG: Fasting plasma glucose. There was no difference in baseline
characteristics between the two groups.

13. 11
At one year, participants in the intervention group were more likely to achieve their lifestyle goals
compared to controls (table 3). Thirty-one percent participants in the intervention group and 54·5%
participants in the control group could not achieve any goal.
Table 3: Number (%) of participants achieving lifestyle goals by allocation group
Number of goals achieved 0 1 2 3 4 P for trend
between
groups
Intervention Group n (%) 42 (31·6) 44 (33·1) 24 (18·0) 15 (11·3) 8 (6·0) <0·001
Control group n (%) 72 (54·5) 45 (34·1) 14 (10·6) 1 (0·8) 0 (0·0)
Those who achieved greater number of lifestyle goals had a greater reduction in weight (figure 3).
Figure 3: Number of goals achieved and corresponding mean weight change in participants by
allocation group at one year

14. 12
Compliance during the first six months was 84·5% (mobile messages: 89·0%, e-mails: 80·0%) but it
dropped to 74·5% at one year (mobile messages: 78·0%, e-mails: 71·0%). Seventy-four (56·0%)
participants were labeled as ‘compliers’ (compliance ≥75%) at one year. The mean weight loss was
significantly greater in the compliers than non-compliers (-2·4 kg vs 0·7 kg, p<0·001). Compliers also
showed greater improvement in awareness scores (27·1% vs 11·9% p<0·001). The intervention was well
received; all participants in the intervention group felt comfortable receiving reminders and found them
useful. Ninety-eight percent participants opted for a continuation of the virtual assistance in future, while
96% would recommend similar interventions for family and friends. There were no adverse events
attributable to the intervention.
Over one year, the direct medical costs of the intervention was INR 2216 ($35·5) per participant in the
control group, and INR 3401 ($54·5) in the intervention group. Thus the incremental cost of reversing
one case of overweight-obesity in one year was INR 10665 ($171) (supplementary material 4).
Discussion
Our findings reveal that there is a high burden of cardio-metabolic risk factors in young Indian IT
employees. Virtual assistance based lifestyle intervention in these high-risk employees was effective at
one year to reduce the prevalence of overweight-obesity. It also encouraged physical activity, healthy
diet practices, and improved awareness levels. There was ~10% drop in overweight-obesity in the
intervention group compared to the control group accompanied by improvements in waist circumference,
total cholesterol, and LDL cholesterol at one year. The participants who achieved larger number of goals
experienced greater weight reduction suggesting an incremental contribution of lifestyle changes. Virtual
assistance through mobile messages and e-mails was an acceptable method to deliver advice and the
participants with higher compliance showed higher benefits.
The observed weight reduction was lower than the intensive interventions used in the clinical efficacy
trials;3,4
but was comparable to that reported in other pragmatic lifestyle interventions.21
In the Diabetes
Prevention Program, every 1 kg of weight loss was associated with a 16% reduction in the risk of
incident diabetes.22
On this background the observed weight reduction may be meaningful, particularly
at a population level. The only Indian trial studying effectiveness of mobile phone messaging on
prevention of diabetes23
reported no significant effect of the intervention on weight, despite lower
cumulative incidence of T2D in the intervention group. Similar to our findings the mobile phone
messaging was found to be an acceptable method to deliver advice and support towards lifestyle
modification.

15. 13
Among the pragmatic lifestyle interventions, there is limited information on the cost-effectiveness of
mobile phone messaging interventions. The Indian Diabetes Prevention Program reported the cost of
intensive lifestyle intervention to prevent one case of diabetes in prediabetic individuals (in 2006) as INR
47341 ($758·7) over three years.7
The cost of virtual assistance based lifestyle intervention in our study
was comparatively low at INR 10665 ($171) for reversing one case of overweight-obesity in a high-risk
population in one year. In the longer term, the costs are likely to be even lower if the intervention is
translated outside the research setting and scaled up.
Given the rapid growth of the IT sector in India and high burden of risk factors in IT professionals, the
need for an effective intervention is obvious. On the background of technology literacy in this sector,
lifestyle advice through mobile phone messages and emails may prove to be an efficient intervention.
Moreover, the number of mobile phone and internet subscribers in urban as well as rural parts of India is
increasing exponentially,24
including amongst children.25
On this background, mobile phone messages
and e-mails can prove encouraging tools for health promotion at population level. These technologies are
affordable, easy to use, and have an instant and wide reach. Their application has been also been tested
in managing several health conditions including management of diabetes,26
although high quality
evidence is confined to adherence to antiretroviral therapy and smoking cessation.27
To our knowledge, this is the first study targeting young high-risk IT employees who were
normoglycemic, unlike the other diabetes prevention trials3–5,23
which enrolled middle-aged participants
with impaired blood glucose. Our study population was a relatively homogenous group exposed to
similar work-environment and socioeconomic status. All the participants were regular users of mobile
and internet services which ensured applicability of the intervention. Both men and women were
included unlike the previous Indian trial23
which enrolled working men alone. We used both mobile
phone messages (information) and e-mails (graphics) as mediums to deliver advice; this combination is
supposedly more retentive than plain text.28
To improve acceptability and compliance, the frequency of
mobile phone messages and e-mails was guided by our preference survey. The screenings and follow-up
assessments were conducted at the worksites considering the hectic schedule of these working
professionals.
One of the limitations is that the trial population consisted of voluntary participants. Such individuals are
likely to have higher awareness and motivation levels, and possibly healthier lifestyles. Given the nature
of our intervention, the participants and field staff could not be masked. However the staff were trained
to follow strict protocols and were supervised closely to minimise bias. The laboratory staff and
statisticians remained masked until the end of the analysis. To reduce the possibility of contamination,
we reminded the participants in the intervention group at every follow-up to maintain confidentiality and

16. 14
not share mobile phone messages and e-mails. The proportion of participants who were lost to follow-up
was similar in both the groups and could be attributed to the high attrition rates in IT industries.20
However, those lost to follow-up were similar to those who remained in terms of baseline characteristics.
Moreover, our analysis is based on intention to treat method. Our study setting was urban, the
application of mobile phone messaging in rural settings needs to be studied. Sustainability of the
intervention and its efficacy over longer period also needs to be studied, we are in the process of setting
up a post trial follow-up.
In summary, we report the effectiveness of a virtual assistance based lifestyle intervention in reducing
overweight and obesity in young Indian IT employees. This low cost intervention could offer a cost-
effective prevention to reduce cardiometabolic risk in this population. Further work is needed to expand
the applications of this novel approach in curtailing the epidemic of obesity, diabetes and cardiovascular
disease.

17. 15
-------------------------------------------------------------------------------------------------------------------
Panel: Research in context
Evidence before this study: We searched systematic reviews, Cocharane reviews, PubMed, and Google
for original studies and reports from Jan 01, 2000 onwards by using a combination of search terms
including ‘mobile phones’, ‘internet’, ‘lifestyle intervention’ ‘cardiometabolic disease’ and ‘diabetes
prevention’. Intensive lifestyle interventions have reported that T2D can be prevented or delayed in
individuals with impaired glycemic status.3–5
Though high quality evidence on use of mobile
interventions is confined to adherence to antiretroviral therapy and smoking cessation;27
its effectiveness
in promoting self management behaviour in individuals with T2D has been reported.26
Application of
mobile phone messaging in prevention of T2D has been tested in only one study.23
So far, the prevention
trials have targeted middle aged adults with impaired glycemic status. Data on cost-effectiveness of
pragmatic trials is scarce especially from developing countries.
Added value of this study: This is the first study targeting young high-risk IT employees at
normoglycemic stage. Mobile phone messages (information) in combination with e-mails (graphics)
were used to promote healthy lifestyle; the combination is thought to be more retentive.28
Technology
literacy in the IT sector ensured applicability of the intervention. Our findings show that a virtual
assistance based lifestyle intervention is effective, cost-effective and acceptable at one year in mitigating
overweight-obesity and other cardiometabolic risk factors associated with T2D.
Implications of all the available evidence: Considering the burgeoning epidemic of T2D and other
cardiometabolic diseases, and an exponential rise in use of mobile and internet services, delivering
advice through such technologies could form an effective primary prevention alternative compared to
more expensive personalized interventions. This approach is potentially scalable at low cost and will
hold important implications for low and middle income countries like India.
-------------------------------------------------------------------------------------------------------------------
Contributors
TL designed and supervised the research, searched the literature, analysed and interpreted the data,
prepared the figures and wrote the manuscript. KK interpreted the data and wrote the manuscript. CJ
analysed and interpreted the data and wrote the manuscript. DB contributed to data collection and
critically reviewed the manuscript. RK contributed to data collection and critically reviewed the
manuscript. AN interpreted the data and critically reviewed the manuscript. CY designed the research
and wrote the manuscript. All authors approved the final version.
Disclosure of interests
All authors declare no competing interests.

18. 16
Acknowledgements
The study was supported by the Department of Science and Technology (DST), New Delhi, India
[DST/INSPIRE Fellowship/2010/(92)] and Diabetes Unit, KEM Hospital Research Centre, Pune. We
thank Deepa Raut, Manisha Deokar, Pooja Jadhav and Shweta Kate for help with data collection.
Virendra Suryawanshi designed the website for the program. We thank the team at Diabetes Unit, KEM
Hospital Research Centre, Pune and Just for Hearts Healthcare Pvt. Ltd., Pune for practical assistance
during the trial. We are grateful to the participants and the management of the IT industries for their
help.

19. 17
References:
1 Misra A, Singhal N, Sivakumar B, Bhagat N, Jaiswal A, Khurana L. Nutrition transition in
India: secular trends in dietary intake and their relationship to diet-related non-communicable
diseases. J Diabetes 2011; 4: 278–92.
2 International Diabetes Federation (IDF). Diabetes Atlas, 6th
Ed. IDF, 2013.
http://www.idf.org/diabetesatlas (accessed May 12, 2014).
3 Tuomilehto J, Lindstrom J, Eriksson JG, et al. Prevention of type 2 diabetes mellitus by changes
in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 2001; 344: 1343–50.
4 Knowler WC, Barrett-Connor E, Fowler SE, et al. Reduction in the incidence of type 2 diabetes
with lifestyle intervention or metformin. N Engl J Med 2002; 346: 393–403.
5 Ramachandran A, Snehalatha C, Mary S, Mukesh B, Bhaskar AD, Vijay V. 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.
6 Hernan WH, Brandle M, Zhang P, et al. Costs associated with the primary prevention of type 2
diabetes mellitus in the diabetes prevention program. Diabetes Care 2003; 26: 36–47.
7 Ramachandran A, Snehalatha C, Yamuna A, Mary S, Ping Z. Cost-effectiveness of the
interventions in the primary prevention of diabetes among Asian Indians: within-trial results of
the Indian Diabetes Prevention Programme (IDPP). Diabetes Care. 2007; 30: 2548–52.
8 India Brand Equity Foundation. IT & ITeS industry in India. Feb 25, 2015.
http://www.ibef.org/industry/information-technology-india.aspx (accessed March 28, 2015).
9 Indian Council for Research on International Economic Relations. Lifestyle diseases hit India's
IT workers. Sept 14, 2007.
http://test.icrier.org/page.asp?MenuID=182&SubCatId=187&SubSubCatId=398 (accessed Oct
25, 2011).
10 American Diabetes Association. Standards of medical care in diabetes-2013. Diabetes Care
2013; 36: S11–66.
11 Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the Joint National Committee on
prevention, detection, evaluation, and treatment of high blood pressure. Hypertension 2003; 42:
1206–52.
12 National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and
Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the
National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and

20. 18
Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report.
Circulation 2002; 106: 3143–421.
13 World Health Organization. Haemoglobin concentrations for the diagnosis of anaemia and
assessment of severity. Vitamin and mineral nutrition information system. Geneva, World
Health Organization, 2011.
14 World Health Organization. Physical status: the use and interpretation of anthropometry. Report
of a WHO Expert Committee. World Health Organ Tech Rep Ser 1995; 854: 1–452.
15 International Diabetes Federation, The IDF consensus worldwide definition of the Metabolic
Syndrome, 2006.
16 Dallal GE. Randomization Plans. August 3, 2007. http://www.randomization.com (accessed
April 13, 2012).
17 Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density
lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem
1972; 18: 499–502.
18 Limaye TY, Wagle SS, Kumaran K, Joglekar CV, Nanivadekar AS, Yajnik CS. Lack of
knowledge about diabetes in Pune - the city of knowledge! Int J Diabetes Dev Ctries DOI
10.1007/s13410-015-0367-3.
19 CDC. How much physical activity do adults need? Center for Disease Control and Prevention.
Dec 1, 2011. http://www.cdc.gov/physicalactivity/everyone/guidelines/adults.html (accessed
May 4, 2012).
20 Technology Business Research. Indian IT sees high attrition as market rebounds: A survey. Feb
27, 2014. http://www.cxotoday.com/story/indian-it-sees-high-attrition-as-market-rebounds
(accessed May 4, 2015).
21 Dunkley AJ, Bodicoat DH, Greaves CJ, et al. Diabetes prevention in the real world:
effectiveness of pragmatic lifestyle interventions for the prevention of type 2 diabetes and of the
impact of adherence to guideline recommendations: a systematic review and meta-analysis.
Diabetes Care 2014; 37: 922–33.
22 Hamman RF, Wing RR, Edelstein SL, Lachin JM, et al. Effect of weight loss with lifestyle
intervention on risk of diabetes. Diabetes Care 2006; 29: 2102–07.
23 Ramachandran A, Snehalatha C, Ram J, 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: 191–98.
24 Telecom Regulatory Authority of India. Highlights on telecom subscription data as on 31st
May
2014. http://trai.gov.in/WriteReadData/WhatsNew/Documents/PR-TSD-May,%2014.pdf
(accessed Oct 18, 2014).

21. 19
25 Groupe Speciale Mobile (GSM) Association and the Mobile Society Research Institute.
Children’s use of mobile phones. An international comparison 2013, Executive Summary. 2014.
http://www.gsma.com/publicpolicy/wp-content/uploads/2014/02/GSMA-NTT-DOCOMO-
Childrens-Report_ExecSummary_RR_Web1.pdf (accessed Oct 30, 2014).
26 Liang X, Wang Q, Yang X, et al. Effect of mobile phone intervention for diabetes on glycaemic
control: a meta-analysis. Diabet Med 2011; 28: 455–63.
27 Vodopivec-Jamsek V, de Jongh T, Gurol-Urganci I, Atun R, Car J. Mobile phone messaging for
preventive health care. Cochrane Database Syst Rev 2012; 12: CD007457.
28 Smiciklas M. The power of infographics, Ed. Wiegand Greg, Pearson education, inc. United
States of America. 2012; 7–16.

22. Supplementary Material
Click here to download Supplementary Material: SUPPLEMENTARY MATERIAL.pdf

23. CONSORT 2010 checklist Page 1
CONSORT 2010 checklist of information to include when reporting a randomised trial
Section/Topic
Item
No Checklist item
Reported
on page No
Title and abstract
1a Identification as a randomised trial in the title 1
1b Structured summary of trial design, methods, results, and conclusions (for specific guidance see CONSORT for abstracts) 2
Introduction
Background and
objectives
2a Scientific background and explanation of rationale 3
2b Specific objectives or hypotheses 3
Methods
Trial design 3a Description of trial design (such as parallel, factorial) including allocation ratio 3
3b Important changes to methods after trial commencement (such as eligibility criteria), with reasons NA
Participants 4a Eligibility criteria for participants 3,4
4b Settings and locations where the data were collected 3
Interventions 5 The interventions for each group with sufficient details to allow replication, including how and when they were
actually administered
5
Outcomes 6a Completely defined pre-specified primary and secondary outcome measures, including how and when they
were assessed
6
6b Any changes to trial outcomes after the trial commenced, with reasons NA
Sample size 7a How sample size was determined 6
7b When applicable, explanation of any interim analyses and stopping guidelines NA
Randomisation:
Sequence
generation
8a Method used to generate the random allocation sequence 4
8b Type of randomisation; details of any restriction (such as blocking and block size) 4
Allocation
concealment
mechanism
9 Mechanism used to implement the random allocation sequence (such as sequentially numbered containers),
describing any steps taken to conceal the sequence until interventions were assigned
4
Implementation 10 Who generated the random allocation sequence, who enrolled participants, and who assigned participants to
interventions
4
*Protocol CONSORT checklist

24. CONSORT 2010 checklist Page 2
Blinding 11a If done, who was blinded after assignment to interventions (for example, participants, care providers, those
assessing outcomes) and how
4
11b If relevant, description of the similarity of interventions NA
Statistical methods 12a Statistical methods used to compare groups for primary and secondary outcomes 6
12b Methods for additional analyses, such as subgroup analyses and adjusted analyses 6
Results
Participant flow (a
diagram is strongly
recommended)
13a For each group, the numbers of participants who were randomly assigned, received intended treatment, and
were analysed for the primary outcome
7,8
13b For each group, losses and exclusions after randomisation, together with reasons 7
Recruitment 14a Dates defining the periods of recruitment and follow-up 3
14b Why the trial ended or was stopped NA
Baseline data 15 A table showing baseline demographic and clinical characteristics for each group 10
Numbers analysed 16 For each group, number of participants (denominator) included in each analysis and whether the analysis was
by original assigned groups
8
Outcomes and
estimation
17a For each primary and secondary outcome, results for each group, and the estimated effect size and its
precision (such as 95% confidence interval)
10
17b For binary outcomes, presentation of both absolute and relative effect sizes is recommended 8
Ancillary analyses 18 Results of any other analyses performed, including subgroup analyses and adjusted analyses, distinguishing
pre-specified from exploratory
11,12
Harms 19 All important harms or unintended effects in each group (for specific guidance see CONSORT for harms) NA
Discussion
Limitations 20 Trial limitations, addressing sources of potential bias, imprecision, and, if relevant, multiplicity of analyses 13,14
Generalisability 21 Generalisability (external validity, applicability) of the trial findings 13
Interpretation 22 Interpretation consistent with results, balancing benefits and harms, and considering other relevant evidence 12,13
Other information
Registration 23 Registration number and name of trial registry 7
Protocol 24 Where the full trial protocol can be accessed, if available NA
Funding 25 Sources of funding and other support (such as supply of drugs), role of funders 7,16