Alive-PD protocol and descriptive paper

Protocol
A Fully Automated Diabetes Prevention Program, Alive-PD:
Program Design and Randomized Controlled Trial Protocol
Gladys Block1,2
, PhD; Kristen MJ Azar3
, RN, MSN, MPH; Torin J Block1
, BA; Robert J Romanelli3
, MPH, PhD;
Heather Carpenter1
, BA; Donald Hopkins1
, MS; Latha Palaniappan3
, MS, MD, FACE, FAHA; Clifford H Block1
,
PhD
1
NutritionQuest, Inc., Berkeley, CA, United States
2
Public Health Nutrition, School of Public Health, University of California, Berkeley, CA, United States
3
Palo Alto Medical Foundation Research Institute, Palo Alto, CA, United States
Corresponding Author:
Gladys Block, PhD
NutritionQuest, Inc.
15 Shattuck Square
Suite 288
Berkeley, CA, 94704
United States
Phone: 1 510 704 8514
Fax: 1 510 704 8996
Email: gblock@berkeley.edu
Abstract
Background: In the United States, 86 million adults have pre-diabetes. Evidence-based interventions that are both cost effective
and widely scalable are needed to prevent diabetes.
Objective: Our goal was to develop a fully automated diabetes prevention program and determine its effectiveness in a randomized
controlled trial.
Methods: Subjects with verified pre-diabetes were recruited to participate in a trial of the effectiveness of Alive-PD, a newly
developed, 1-year, fully automated behavior change program delivered by email and Web. The program involves weekly tailored
goal-setting, team-based and individual challenges, gamification, and other opportunities for interaction. An accompanying mobile
phone app supports goal-setting and activity planning. For the trial, participants were randomized by computer algorithm to start
the program immediately or after a 6-month delay. The primary outcome measures are change in HbA1c and fasting glucose
from baseline to 6 months. The secondary outcome measures are change in HbA1c, glucose, lipids, body mass index (BMI),
weight, waist circumference, and blood pressure at 3, 6, 9, and 12 months. Randomization and delivery of the intervention are
independent of clinic staff, who are blinded to treatment assignment. Outcomes will be evaluated for the intention-to-treat and
per-protocol populations.
Results: A total of 340 subjects with pre-diabetes were randomized to the intervention (n=164) or delayed-entry control group
(n=176). Baseline characteristics were as follows: mean age 55 (SD 8.9); mean BMI 31.1 (SD 4.3); male 68.5%; mean fasting
glucose 109.9 (SD 8.4) mg/dL; and mean HbA1c 5.6 (SD 0.3)%. Data collection and analysis are in progress. We hypothesize
that participants in the intervention group will achieve statistically significant reductions in fasting glucose and HbA1c as compared
to the control group at 6 months post baseline.
Conclusions: The randomized trial will provide rigorous evidence regarding the efficacy of this Web- and Internet-based
program in reducing or preventing progression of glycemic markers and indirectly in preventing progression to diabetes.
Trial Registration: ClinicalTrials.gov NCT01479062; http://clinicaltrials.gov/show/NCT01479062 (Archived by WebCite at
http://www.webcitation.org/6U8ODy1vo).
(JMIR Res Protoc 2015;4(1):e3) doi:10.2196/resprot.4046
KEYWORDS
prediabetes; insulin resistance; diabetes; prevention; obesity; physical activity; internet; world wide web; blood glucose; metabolic
syndrome
JMIR Res Protoc 2015 | vol. 4 | iss. 1 | e3 | p.1http://www.researchprotocols.org/2015/1/e3/
(page number not for citation purposes)
Block et alJMIR RESEARCH PROTOCOLS
XSL•FO
RenderX

Introduction
In the United States, 86 million adults, more than one-third of
the population, have pre-diabetes [1]. It has been estimated that
over half will eventually progress to type 2 diabetes [2] unless
they make changes to their dietary and physical activity
behaviors. Therefore, there is an urgent need to reach large
numbers of individuals with pre-diabetes with strategies to help
them produce long-term behavioral changes and weight
management.
Previous interventions intended to prevent diabetes have
incorporated various levels and combinations of lifestyle
coaching, in-person interaction, and technology-enhanced tools.
These various models differ in reach, cost, and effectiveness.
Intensive personal and group coaching can produce medically
significant changes in those persons with pre-diabetes who
choose to participate, as demonstrated by the Diabetes
Prevention Program (DPP) [3]. While efficacious, such
interventions are costly [4] and challenging to implement [5].
Adaptations of the DPP model in community settings [5-7] that
rely on group meetings and frequent in-person contact require
substantial organizational skills and professional or
semi-professional implementation [8]. Equally important is the
reluctance of many, even those quite highly motivated, to
dedicate the time to attend a series of onsite meetings. Costs
remain substantial [4,8] and effectiveness modest [5,9-11].
Given the well-established transience of weight loss for most
people, interventions that can produce robust establishment of
new habits and that also permit long-term reinforcement and
support in the future are called for. Internet-delivered
applications, if proven sufficiently effective, could meet a
substantial part of that need. A number of such efforts have
been described [9,12-16]. A meta-analysis [9] found that among
electronic media-assisted programs, there was a statistically
significant but modest mean weight loss of 4.2% body weight.
However, often such programs rely on some form of
contribution by human counselors in addition to the
technological component, which substantially increases cost
and will prevent the large economies of scale that a fully
automated system would provide.
Fully automated technological approaches have been found to
improve diet and/or physical activity behaviors [17,18]. Alive!,
which forms the basis of the diabetes prevention program
described here, was shown in a randomized trial to produce
significant improvements in physical activity and diet [19,20].
However, fully automated or nearly fully automated programs
have been studied only rarely for effectiveness on weight loss
[12], and almost not at all for diabetes prevention.
To address the need for a diabetes prevention program that can
reach thousands or millions of pre-diabetics with an
evidence-based intervention, we designed the
Alive-PreventDiabetes (Alive-PD) program. Alive-PD is a
newly developed, fully automated, 1-year behavior change
program to prevent the progression of pre-diabetes to diabetes.
Like its precursor program [19,20], it emphasizes lasting
lifestyle change in diet and physical activity, which can lead to
improvements in blood sugar and weight. It is a stand-alone
intervention, although it can also be used to supplement clinician
recommendations or to provide long-term support following
pre-diabetes classes. This paper describes the Alive-PD program
and describes the protocol of the randomized trial currently
under way to evaluate its efficacy.
Methods
Study Design Overview
The Alive-PD study is a randomized parallel-group controlled
trial of the Alive-PD program versus a 6-month-delayed control
group among participants with pre-diabetes. The primary
outcome measures are change in hemoglobin A1c (HbA1c) and
fasting glucose after 6 months. After the study began, the
follow-up period was extended to 12 months for both arms. We
hypothesize that those in the intervention group will achieve
statistically significantly greater reductions in fasting glucose
and HbA1c than a delayed-entry control group at 6 months post
baseline and that the intervention group will retain a statistically
significant change at 1 year. Clinic assessments and laboratory
data are collected at 3, 6, 9, and 12 months. The trial is ongoing.
This randomized clinical trial is being conducted in collaboration
with the Palo Alto Medical Foundation Research Institute
(PAMFRI). The study design and materials were approved by
the Institutional Review Boards of NutritionQuest and of the
Palo Alto Medical Foundation (PAMF). See Multimedia
Appendix 1 for the CONSORT-EHEALTH checklist [21].
Participant Eligibility Criteria
Participants were recruited from a primary care health center
of PAMF, a community-based multi-specialty group practice
in northern California. We used the following inclusion criteria
for participation: primary care patients with pre-diabetes, age
30-69 years old, BMI ≥27 (BMI >25 for Asian race/ethnic
subgroups only), fasting glucose measure in the pre-diabetic
range (100-125 mg/dL) or HbA1c in the pre-diabetic range
(5.7-6.4%) or physician diagnosis of pre-diabetes, HbA1c ≥5.3%
and fasting glucose ≥95 mg/dL (added after study under way),
a visit to PAMF primary care department within the past 5 years,
access to email and Internet, and able to read and understand
English. Those meeting the following exclusion criteria were
not eligible: presence of medical conditions contraindicating
gradual adoption of moderate physical activity, current use of
diabetes medications or weight loss medications, currently
pregnant or planning to become pregnant during the period of
the study, currently doing more than 150 minutes/week of
moderate/vigorous physical activity and currently on a
low-carbohydrate diet (by self-report), and current participation
in another clinical trial.
Procedures
Recruitment and Screening Process
Potential participants were pre-screened and identified through
an electronic health records (EHR) query using broad eligibility
criteria of a recent glucose or HbA1c within the pre-diabetes
range, in addition to other demographic criteria (Figure 1). In
addition to the EHR query, posters were placed in the lobby of
XSL•FO
RenderX

the PAMF clinics. Study invitation letters were mailed to
potentially eligible participants, with phone follow-up to explain
the study, conduct further eligibility assessment, and invite
participation. Those who remained eligible and were interested
in participating were invited to attend the baseline visit at the
PAMF Research Institute for confirmation of eligibility based
on point-of-care laboratory testing for glucose and HbA1c.
Subjects were enrolled in the study between February and June
2014.
Figure 1. Flowchart of participant screening, recruitment, and randomization.
Baseline Assessment and Clinic Visit
At the baseline clinic visit, HbA1c, fasting glucose, total and
high-density lipoprotein (HDL) cholesterol, triglycerides, and
low-density lipoprotein (LDL) by calculation were assessed
through point-of-care whole blood samples (Siemens DCA
Vantage point-of-care analyzer [22] and Alere Cholestech LDX
point-of-care analyzer [23] respectively). In addition to meeting
other eligibility criteria, subjects were deemed eligible if either
HbA1c and/or fasting glucose fell within the pre-diabetic range,
and neither measure reached the diabetic range. A number of
subjects who were eligible by one measure were found to have
quite low HbA1c or fasting glucose; a further restriction, HbA1c
≥5.3% and glucose ≥95, was added after the study had begun.
Prospective participants with one measure in the diabetes range
and the other in the pre-diabetes range were deemed
provisionally eligible pending physician review and verification
of pre-diabetes status. Signed informed consent was obtained.
Other biometric measures included blood pressure, height,
weight, and waist circumference, measured according to the
procedures recommended by the Airlie Conference. In addition,
a baseline questionnaire was administered, assessing medication
XSL•FO
RenderX

use, adverse events and participation in activities relevant to
weight loss, diet, and physical activity. With separate consent,
a blood sample was obtained by FTA (Flinders Technology
Associates) card for future research on genetic analysis of the
intervention’s efficacy and disease progression.
Once enrolled and confirmed eligible, subjects received brief
(5-10 minutes) instruction that they were at risk for developing
diabetes and that increased physical activity as well as changes
in their dietary behaviors could help prevent progression to
diabetes. Research staff assisted participants in signing into an
account for the Alive-PD Web-based program, to confirm to
the Alive-PD system that they had completed the informed
consent and to provide their email address and password to the
system. All subsequent communications from and interactions
with the electronic Alive-PD program took place outside of the
clinic.
Subsequent Clinic Visits
At 3, 6, 9, and 12 months, the same biometric and questionnaire
measures are repeated. An adverse events questionnaire is also
administered at each follow-up.
Randomization and Blinding
After the baseline clinic visit and the initial account set-up,
randomization and delivery of the intervention is conducted
electronically by the Alive-PD program, completely independent
of clinic staff. After leaving the clinic, participants are sent an
introductory email from the automated Alive-PD system, with
further information and questionnaires. These and all further
online interactions are done at a time and place of the
participants’ choosing. An online baseline questionnaire is
required, to provide demographic and BMI information required
for randomization and to provide brief eating habits and physical
activity data for baseline comparison of intervention and control
groups. Participants are randomized by computer algorithm,
with blocked stratified randomization to achieve balance on
sex, BMI (above/below 35), and race/ethnicity (white
non-Hispanic, other). Participants are informed by the system
by email about whether they have been randomized to begin
the program immediately or after a 6-month delay. Those in the
delayed group receive no further contact until a reminder to
complete a 3-month and 6-month online follow-up
questionnaire. Those randomized to begin immediately receive
a more extensive diet and activity questionnaire.
Reimbursement
Participants are reimbursed for attending the clinic and providing
the biometric outcome data. They receive a US $25 gift card at
the 3-month clinic visit and a US $50 gift card at the 6-month
clinic visit. If they re-consent to participate in the study
extension, they receive a US $25 gift card at the 9-month clinic
visit and a US $50 gift card at the 12-month clinic visit.
Intervention
Overview
Alive-PD is a fully automated intervention, with no human
coaching or advice. It is delivered via an individualized website
and interactive emails and is supplemented by a mobile app and
automated phone and print modules. Participants complete a
diet and physical activity assessment online at baseline, followed
by immediate automated individualized feedback. They also
complete an activity planning tool to guide improvements in
aerobic activity. Participants then engage in a program of weekly
tailored goal setting and tracking over the next 6 months and
every other week for the subsequent 6 months. Participants
work simultaneously on dietary and physical activity behaviors.
Targeted behavior changes include increasing physical activity,
decreasing added sugars and some starchy carbohydrates,
decreasing saturated and trans fats, increasing fruit and vegetable
consumption, and achieving modest weight loss or weight
maintenance. Participants are invited (but not required) to choose
a long-term weight loss target of 5%, 7%, or 10% of body
weight. Changes in food type and reduction in portion size are
emphasized as a means of reducing calorie intake, rather than
specific calorie targets or counting. The focus throughout is on
the objective of lowering hemoglobin A1c and preventing
diabetes. See Figure 2 and Multimedia Appendix 2 for
screenshots and other information.
XSL•FO
RenderX

Figure 2. Alive-PD personal home page. Users are taken here when they choose goals in the weekly email. Or they may go here directly at any time
-- for example, to log, check or send messages, get automated coaching. Further down the page are infographics and quizzes. If they hadn't yet chosen
or reported on goals this week, the home page would show the week's suggested goals to choose, and request a report on last week's goals.
Features of the Alive-PD Program and Website
The program begins with an online baseline questionnaire that
assesses the individual’s intake of added sugars, carbohydrates,
saturated and trans fats, fruits and vegetables, and time spent
in moderate and vigorous intensity activity, walking, sedentary
activities and strength training. Questionnaire content is tailored
to the participant’s reported ethnic dietary habits. Analysis
occurs with each response, allowing for continuous and relevant
feedback on the individual’s reported diet and activity, in
relation to national guidelines and type 2 diabetes risk factors.
The questionnaire responses provide the basis for tailoring
specific goals and tips for changes in diet, physical activity, and
weight control that the program will recommend to the
participant throughout the program. Questions on lifestyle, diet,
and physical activity preferences, and psychosocial
characteristics form the basis of additional tailoring.
Weekly tailored goal setting is one of the key behavioral features
of Alive-PD. In addition to long-term goal setting, individually
relevant small-step goals are suggested weekly, developed by
an algorithm that draws from the participant’s questionnaire
responses, the participant-constructed activity plan, and from
an extensive database of potential goals. For example,
participants with added sugar intake above the recommended
level are offered weekly goals targeting the individual’s own
reported sources of added sugars. The goal recommendation
algorithm adjusts in response to past performance.
The system suggests to participants that they select one eating
habits goal and one physical activity goal from among those
suggested each week. However, they may select any number
of goals and/or may write their own goals each week. The
program also prompts them each week to report on their success
at the previous week’s goals. The number and type of successful
goals is tracked for the respondent. Mid-week reminders about
the goals chosen that week serve to keep the participant engaged
and the overall objective salient. Participants who have not
responded by mid-week by choosing a goal are reminded by
email and/or mobile phone of their self-reported motivations
for joining the program and are encouraged again to commit to
1-2 goals for the remainder of the week.
An interactive activity planning tool forms the basis of physical
activity goal setting. Alive-PD automatically creates a
recommended activity plan based on a participant’s measured
activity level, but the participant may edit the plan at any time.
The default target is 150 minutes per week, but those already
achieving that level may aim higher. As participants interact
with each setting, their activity plan is instantly recomputed and
drawn onscreen. Their saved plan guides the tailored aerobic
activity goals recommended to the participant each week.
XSL•FO
RenderX

Two weekly Health Notes, developed by Certified Diabetes
Educators and behavioral experts, provide health information.
Topics include subjects such as the role of diet and physical
activity in preventing diabetes and other chronic diseases,
physical activity guidelines, dietary sources of saturated fat and
sugars, portion control, and behavioral strategies including
self-monitoring, mindfulness, stress reduction, and sleep. Each
Health Note includes a quiz, and participants may rate the Health
Note and share it on social media. Delivery of the Health Notes
may be tailored to the ethnicity of the participant.
The program promotes social support through a team system,
which allows participants to send messages to each other with
either a pre-drafted motivational message or a personalized
message of support. The system also encourages participants
to elicit the support of their personal social network. Participants
may invite friends and family to the program, and those who
accept receive weekly emails from the system, which include
the Health Notes the participant is reading that week and a
function to send a message to the participant’s Alive-PD inbox.
Other features include a Daily Tracker to log weight, activity,
and dietary intake, with automatic graphing over time; an
automated coaching tool to help participants overcome common
barriers; weekly infographics that reinforce core messages of
the program; downloadable worksheets; and links to external
resources.
Program engagement strategies combine intrinsic and extrinsic
motivators, with individual and optional team aspects and
software gamification features. At the individual level,
interaction with central components of the intervention is
incentivized using a points system. Participants are awarded
points for the following behaviors: setting and reporting goals;
successfully accomplishing goals; reading Health Notes and
taking the related quizzes; logging body weight, activity, or
dietary intake; and sending support messages to other Alive-PD
participants. Points can be redeemed for modest monetary
rewards or may be contributed to a pooled donation to the
American Diabetes Association. At the team level, teams
compete for small monetary rewards based on the greatest
average HbA1c reduction in each 3-month period.
Automated Phone and Print
The content of these Alive-PD modules was developed by
nutrition and physical activity professionals, through
collaboration with Stanford and Brown University experts. With
these technologies, individually tailored print and phone
messages are delivered automatically by computer algorithm.
They make use of the participant’s baseline responses to factors
such as barriers and motivations for behavior change, participant
real-time dropout or adherence status, and ongoing participant
reports on success at goals undertaken in the Alive-PD program.
Individually tailored printed materials are sent monthly.
Automated individually tailored phone coaching is delivered
every 2 weeks via Interactive Voice Response technology (IVR).
Participant responses to the IVR questions are recorded in the
system and used to inform subsequent IVR interactions. Both
phone and print are entirely automated and require no real-time
professional or semiprofessional input.
Alive-PD Mobile App
Alive-PD includes an Android and iPhone app that extends
some of the program functionality to the participant’s mobile
phone. The app allows participants to select Alive-PD weekly
goals, report on accomplishment of previous goals, and set
calendar reminders. The program also reinforces selected goals
via push notifications and coaxes potential dropouts to engage
with the program. The mobile phone app was introduced after
approximately a quarter of subjects had begun the trial, at which
time its availability was announced to all participants.
Online Assessment of Non-Physiologic Outcomes
Physiologic outcomes, which are the primary outcomes, are
assessed by anthropometric or biometric measures in the clinic,
as stated above. Other outcomes are assessed by
self-administered questionnaire or through automated capture
of use metrics. These include factors such as number of weeks
in which a goal was chosen, number of goals chosen, number
of goals achieved, number of team support messages sent, and
number of quizzes answered. Changes in physical activity and
eating habits are assessed by online modified Block
questionnaire. Satisfaction with the program is also assessed
by online questionnaire.
Theoretical Basis
Alive-PD’s intervention strategies are based on principles
derived from several models and bodies of behavior change
research. The basic objective, derived from learning theory and
other habit formation research [24-26] is to have participants
gradually incorporate new eating and physical activity behaviors
into their daily lives until the behaviors are both habitual and
substantial enough to reduce diabetes risk. A focus on small,
achievable, personally relevant goals provides continued reward
from accomplishing those goals.
Strategies and features of the program are consistent with several
other bodies of research, including models centering on cues
and triggers [27,28], social cognitive theory [29,30], and the
theory of planned behavior [31]. Concepts from behavioral
economics [28,32] underlie suggestions for making desired
behaviors more convenient and undesired behaviors less
convenient. In addition, strategies have been influenced by
research on willpower showing the neurophysiological limits
of sustained cognitive control for behaviors like dieting [33].
Several empirically validated positive psychology methods
[34,35] promote increased resilience and optimism, helpful
when undertaking major changes. Techniques based on
mindfulness research [36,37] can provide support in the face of
cravings. The program’s continuing flow of information on the
consequences of diabetes, the ability to reduce one’s risk and
skills for doing so is consistent with the health belief model
[38]. Promotion of social support through the messaging system,
teams, and inclusion of real-world friends and family can
provide reinforcement of desirable changes, buffering of stress,
and continuing engagement. Further maximizing of engagement
through elements such as gamification provides continuing
salience of issues around health, nutrition, and physical activity.
In sum, techniques consistent with a variety of theories and
XSL•FO
RenderX

models provide support for individual differences and for
individuals at different times, as they gradually adopt new habits.
Measures and Data Collection
The primary outcome of Alive-PD is change in HbA1c and
fasting glucose at 6 months from baseline, assessed through
point-of-care whole blood samples (see Baseline Assessment
and Clinic Visit, above). Secondary endpoints are
clinic-measured changes in other biometric measures (weight,
waist circumference, lipids and other metabolic syndrome
factors, and blood pressure). We will examine retention of
changes at 12 months in a pre-post analysis. In addition, we
will examine self-reported behavior changes in eating and
physical activity.
Statistical Analysis
Sample Size
Sample size was estimated using data from Davis et al [39],
who conducted a dietary intervention in patients with diabetes.
They reported standard deviations for change in HbA1c of
0.9-1.4% for one arm of the study. With an SD of 1.4 and an
alpha of .05, a final sample of 268 participants would provide
80% power to detect a minimum detectable difference in change
for HbA1c of 0.48. We planned to enroll 314 persons to achieve
a completed sample of 268 after a 15% dropout rate.
Data Analysis
The primary approach for analyses of treatment effects on
glycemic markers (HbA1c and glucose) and other biometric
markers will be on the intention-to-treat population. Variables
missing at follow-up will be imputed using a maximum
likelihood approach. We will also repeat analyses on the
per-protocol population, in which effects will be assessed among
those providing follow-up biometric data and in relation to the
participants’degree of interaction with the program. Outcomes
will be analyzed using linear regression models. Change in the
outcome measures (HbA1c, glucose, or weight) will be included
in regression models as the dependent variable, and treatment
group will be included as the primary predictor variable
(intervention vs delayed control). The baseline value of the
outcome measure of interest will be included as a covariate.
Normality, interaction, and potential confounding will also be
investigated, and potential confounders will be included as
appropriate. We will also examine the proportion of participants
who achieve and retain categories of body weight loss, and the
proportion who move from pre-diabetic to normal HbA1c. These
binary outcomes will be modeled in logistic regression models
with an approach similar to that described for linear regressions.
Clinic staff obtaining the biometric measurements are blinded
to treatment group and data analysis is conducted on masked
data.
Results
Enrollment and Baseline Characteristics
A total of 349 participants completed the Informed Consent,
and 340 completed the baseline online questionnaire and were
randomized to Alive-PD or 6-month delayed control. The 9
participants who did not complete the baseline questionnaire
did not differ in age, sex, race, BMI, or lipids, but did have a
significantly lower fasting glucose but higher HbA1c (data not
shown).
Baseline characteristics are shown in Table 1. The large
proportion of males (68.5%, 233/340) is notable, and quite the
opposite of what is commonly seen in health studies. It is also
notable that only 45% of the sample were in the pre-diabetic
range by HbA1c and that the mean HbA1c is in the normal
range.
XSL•FO
RenderX

Table 1. Baseline demographics and clinical characteristics.
P a
Intervention (n=164)Control (n=176)All (n=340)CategoryVariable
.7755.1 (8.8)54.8 (9.1)55.0 (8.9)Age, years, mean (SD)
.7553 (32.2)54 (30.7)107 (31.5)Female, n (%)
.38Race/ethnicity, n (%) b
100 (61.0)104 (59.1)204 (60.0)White
6 (3.7)1 (0.6)7 (2.1)Black
27 (16.5)36 (20.4)63 (18.5)Asian
7 (4.3)9 (5.1)16 (4.7)Hispanic
3 (1.8)5 (2.8)8 (2.3)Other
21 (12.8)21 (11.9)42 (12.4)Not reported
Blood pressure, mmHg, mean (SD)
.99130.4 (15.0)130.4 (14.5)130.4 (14.7)Systolic
.4582.0 (8.1)82.6 (8.6)82.3 (8.4)Diastolic
.64203.9 (32.9)205.6 (36.5)204.8 (34.8)Weight, lbs, mean (SD)
.8631.1 (4.4)31.2 (4.3)31.1 (4.3)BMI, kg/m2
, mean (SD)
.59102.5 (10.4)103.1 (11.2)102.8 (10.8)Waist circumference, cm, mean (SD)
.52110.2 (8.6)109.0 (8.3)109.9 (8.4)Glucose, mg/dL, mean (SD)
.945.6 (0.3)5.6 (0.3)5.6 (0.3)HbA1c, %, mean (SD)
.83191.2 (30.2)191.9 (33.0)191.5 (31.7)Total cholesterol, mg/dL, mean (SD)
.72115.5 (27.3)116.5 (27.9)116.0 (27.6)LDL cholesterol, mg/dLc
, mean (SD)
.4047.8 (14.5)46.5 (13.0)47.3 (14.0)HDL cholesterol, mg/dLc
, mean (SD)
.24139.4 (67.3)148.2 (68.8)144.6 (73.7)Triglycerides, mg/dLc
, mean (SD)
a
Significance of difference between intervention and control.
b
Native American/Alaskan, Native Hawaiian/Pacific Islander, reported as multiple races or “other”. Ethnicity from EHR.
c
For LDL: Control, n=172, Intervention, n=161; for HDL: Control, n=176, Intervention, n=162; for triglycerides: Control, n=174, Intervention, n=162;
for waist circumference: Control, n=175, Intervention, n=163.
Preliminary Outcome Results
Preliminary data on treatment-control differences indicate a
statistically significant treatment effect on HbA1c, fasting
glucose, and body weight (data reported elsewhere). These
significant treatment effects were seen despite the fact that the
control group achieved a statistically significant (P<.01)
reduction in HbA1c and weight.
Preliminary Participation, Retention Results
Dropout rate (defined as permanently declining to attend further
clinic visits) as of the 3-month clinic visit is 5.9% (20/340).
Also as of the 3-month time point, 78% of the intervention group
were still participating actively in the online program, by
choosing or reporting on a goal, logging weight or activity,
answering a quiz, or sending a message to team members via
the online messaging system.
Discussion
Summary
The Alive-PD program is one of very few interventions designed
as a fully automated algorithm-driven program for persons at
risk of developing diabetes. If proven effective in the
randomized trial described here, it could reach many of the 86
million US pre-diabetics with an evidence-based
behavior-change intervention. In addition, the delivery of the
program is highly configurable and can be readily adapted for
other medical conditions or for populations with other eating
and activity behaviors. Alive-PD will be commercially available.
Limitations
We will not have the duration or sample size to have a diabetes
incidence endpoint. However, we do have the power to assess
efficacy for glycemic control and weight loss. Furthermore, the
fact that the study is restricted to participants with email and
Internet access is a limitation. However, as of August 2011,
92% of American adults used email and the Internet, including
93% of whites, 87% of African Americans, and 88% of
XSL•FO
RenderX

Hispanics, and over 50% had a mobile phone. Finally, we
recognize that wholly automated interactions, without
person-to-person contact, can be limiting for many. However,
such a system can reach large numbers at a much lower cost
than systems requiring interaction with human coaches. Even
modest changes on a proportion of the very large target
population could result in major benefits in cost savings and
disease prevention. If only 5% of the 86 million Americans with
pre-diabetes were reached with an effective automated
intervention program, and as few as one-third of them succeeded
in maintaining or reducing their pre-diabetic status, that would
mean more than 1 million Americans with reduced risk of type
2 diabetes and billions of dollars saved in health care costs.
Conclusions
Alive-PD is a multi-channel automated intervention to prevent
diabetes by lowering or preventing the increase in HbA1c and
glucose. If proven effective, Alive-PD would demonstrate how
a preventative intervention can be provided with wide reach,
high fidelity, and low cost for persons with pre-diabetes.

Acknowledgments
Research reported in this publication was supported by the National Institute of Nursing Research of the National Institutes of
Health under Award Number R44NR012617. The content is solely the responsibility of the authors and does not necessarily
represent the official views of the National Institutes of Health.
Authors' Contributions
GB obtained funding and led the study. TB, CB, GB, and HC designed and implemented the Alive-PD program. DH conducted
and led the systems engineering. KA led the sub-award to PAMFRI, RR was co-investigator of the PAMFRI sub-award, and both
conducted the recruitment, biological measurements, and clinic-based data collection. LP facilitated the establishment of the
sub-award. GB, CB, TB, RR, KA, and LP contributed to the drafting of the manuscript.
Conflicts of Interest
CB, TB, and GB are co-owners of NutritionQuest, which developed the Alive-PD program.
Multimedia Appendix 1
CONSORT-EHEALTH checklist V1.6.1 [21].
[PDF File (Adobe PDF File), 948KB - resprot_v4i1e3_app1.pdf ]
Screenshots, Medicine 2.0 2014 talk, Gladys Block.
[PPT File (Microsoft PowerPoint Presentation), 4MB - resprot_v4i1e3_app2.ppt ]
NIH reviewers' comments.
[PDF File (Adobe PDF File), 77KB - resprot_v4i1e3_app3.PDF ]
References
1. American Diabetes Association. 2014. Statistics About Diabetes URL: http://www.diabetes.org/diabetes-basics/statistics/
[accessed 2014-11-14] [WebCite Cache ID 6U4Tx9Xxu]
2. Twigg SM, Kamp MC, Davis TM, Neylon EK, Flack JR, Australian Diabetes Society, Australian Diabetes Educators
Association. Prediabetes: a position statement from the Australian Diabetes Society and Australian Diabetes Educators
Association. Med J Aust 2007 May 7;186(9):461-465. [Medline: 17484708]
3. Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, Walker EA, Diabetes Prevention Program Research
Group. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 2002 Feb
7;346(6):393-403 [FREE Full text] [doi: 10.1056/NEJMoa012512] [Medline: 11832527]
4. Lawlor MS, Blackwell CS, Isom SP, Katula JA, Vitolins MZ, Morgan TM, et al. Cost of a group translation of the Diabetes
Prevention Program: Healthy Living Partnerships to Prevent Diabetes. Am J Prev Med 2013 Apr;44(4 Suppl 4):S381-S389
[FREE Full text] [doi: 10.1016/j.amepre.2012.12.016] [Medline: 23498303]
5. Ackermann RT, Finch EA, Brizendine E, Zhou H, Marrero DG. Translating the Diabetes Prevention Program into the
community. The DEPLOY Pilot Study. Am J Prev Med 2008 Oct;35(4):357-363 [FREE Full text] [doi:
10.1016/j.amepre.2008.06.035] [Medline: 18779029]
XSL•FO
RenderX

6. Whittemore R, Melkus G, Wagner J, Dziura J, Northrup V, Grey M. Translating the diabetes prevention program to primary
care: a pilot study. Nurs Res 2009;58(1):2-12 [FREE Full text] [doi: 10.1097/NNR.0b013e31818fcef3] [Medline: 19092550]
7. Davis-Smith YM, Davis-Smith M, Boltri JM, Seale JP, Shellenberger S, Blalock T, et al. Implementing a diabetes prevention
program in a rural African-American church. J Natl Med Assoc 2007 Apr;99(4):440-446. [Medline: 17444435]
8. Vojta D, Koehler TB, Longjohn M, Lever JA, Caputo NF. A coordinated national model for diabetes prevention: linking
health systems to an evidence-based community program. Am J Prev Med 2013 Apr;44(4 Suppl 4):S301-S306. [doi:
10.1016/j.amepre.2012.12.018] [Medline: 23498291]
9. Ali MK, Echouffo-Tcheugui J, Williamson DF. How effective were lifestyle interventions in real-world settings that were
modeled on the Diabetes Prevention Program? Health Aff (Millwood) 2012 Jan;31(1):67-75 [FREE Full text] [doi:
10.1377/hlthaff.2011.1009] [Medline: 22232096]
10. Seidel MC, Powell RO, Zgibor JC, Siminerio LM, Piatt GA. Translating the Diabetes Prevention Program into an urban
medically underserved community: a nonrandomized prospective intervention study. Diabetes Care 2008 Apr;31(4):684-689.
[doi: 10.2337/dc07-1869] [Medline: 18252904]
11. Kramer MK, Kriska AM, Venditti EM, Semler LN, Miller RG, McDonald T, et al. A novel approach to diabetes prevention:
evaluation of the Group Lifestyle Balance program delivered via DVD. Diabetes Res Clin Pract 2010 Dec;90(3):e60-e63.
[doi: 10.1016/j.diabres.2010.08.013] [Medline: 20863586]
12. Thomas JG, Leahey TM, Wing RR. An Automated Internet Behavioral Weight Loss Program by Physician Referral: A
Randomized, Controlled Trial. Diabetes Care 2014 Nov 17. [doi: 10.2337/dc14-1474] [Medline: 25404659]
13. McTigue KM, Conroy MB, Hess R, Bryce CL, Fiorillo AB, Fischer GS, et al. Using the internet to translate an evidence-based
lifestyle intervention into practice. Telemed J E Health 2009 Nov;15(9):851-858. [doi: 10.1089/tmj.2009.0036] [Medline:
19919191]
14. McCoy MR, Couch D, Duncan ND, Lynch GS. Evaluating an internet weight loss program for diabetes prevention. Health
Promot Int 2005 Sep;20(3):221-228 [FREE Full text] [doi: 10.1093/heapro/dai006] [Medline: 15797900]
15. Hunter CM, Peterson AL, Alvarez LM, Poston WC, Brundige AR, Haddock CK, et al. Weight management using the
internet a randomized controlled trial. Am J Prev Med 2008 Feb;34(2):119-126. [doi: 10.1016/j.amepre.2007.09.026]
[Medline: 18201641]
16. Sepah SC, Jiang L, Peters AL. Translating the Diabetes Prevention Program into an Online Social Network: Validation
against CDC Standards. Diabetes Educ 2014 Apr 10;40(4):435-443. [doi: 10.1177/0145721714531339] [Medline: 24723130]
17. Oenema A, Brug J, Dijkstra A, de Weerdt I, de Vries H. Efficacy and use of an internet-delivered computer-tailored lifestyle
intervention, targeting saturated fat intake, physical activity and smoking cessation: a randomized controlled trial. Ann
Behav Med 2008 Apr;35(2):125-135. [doi: 10.1007/s12160-008-9023-1] [Medline: 18363076]
18. van den Berg MH, Schoones JW, Vliet Vlieland TP. Internet-based physical activity interventions: a systematic review of
the literature. J Med Internet Res 2007;9(3):e26 [FREE Full text] [doi: 10.2196/jmir.9.3.e26] [Medline: 17942388]
19. Sternfeld B, Block C, Quesenberry CP, Block TJ, Husson G, Norris JC, et al. Improving diet and physical activity with
ALIVE: a worksite randomized trial. Am J Prev Med 2009 Jun;36(6):475-483. [doi: 10.1016/j.amepre.2009.01.036]
[Medline: 19460655]
20. Block G, Sternfeld B, Block CH, Block TJ, Norris J, Hopkins D, et al. Development of Alive! (A Lifestyle Intervention
Via Email), and its effect on health-related quality of life, presenteeism, and other behavioral outcomes: randomized
controlled trial. J Med Internet Res 2008;10(4):e43 [FREE Full text] [doi: 10.2196/jmir.1112] [Medline: 19019818]
21. Eysenbach G, CONSORT-EHEALTH Group. CONSORT-EHEALTH: improving and standardizing evaluation reports of
Web-based and mobile health interventions. J Med Internet Res 2011;13(4):e126 [FREE Full text] [doi: 10.2196/jmir.1923]
[Medline: 22209829]
22. Zercher A, Schulman SRL, Boone J. Quantitative measurement of hemoglobin A1c on the DCA Vantage Point-of-Care
Analyzer as a diagnostic test for diabetes: An internal validation study. URL: http://www.healthcare.siemens.com/
siemens_hwem-hwem_ssxa_websites-context-root/wcm/idc/groups/public/@global/@lab/@poc/documents/download/
mday/nzg2/~edisp/130890-gc1_dca_hba1c_precision_study_white_paper_to_support_ous_dx_claim_final_web-01360756.
pdf [accessed 2014-11-15] [WebCite Cache ID 6U6r7fHe3]
23. Cholestech Corporation. 2005. Accuracy of a rapid, finger-stick lipid profile method is comparable to commercial laboratory
methods URL: http://www.hmscweb.com/PDF_Files/Cholestech/
Technical_Brief_LDX_-_Lipid_Profile_accuracy_and_precision.pdf [accessed 2014-11-15] [WebCite Cache ID 6U6rQUlwp]
24. Duhigg C. Power of Habit: Why We Do What We Do, and How to Change. New York: Random House; 2012.
25. Hull CL. Essentials of behavior. Westport, CT: Greenwood Press; 1951.
26. Dickinson A. Actions and Habits: The Development of Behavioural Autonomy. Philosophical Transactions of the Royal
Society B: Biological Sciences 1985 Feb 13;308(1135):67-78. [doi: 10.1098/rstb.1985.0010]
27. Fogg GJ. Stanford Persuasive Tech Lab. 2014. URL: http://captology.stanford.edu/ [accessed 2014-11-14] [WebCite Cache
ID 6U4ULymv1]
28. Wansink B. Mindless Eating: Why We Eat More Than We Think. New York: Bantam; 2006.
29. Brownell KD, Marlatt GA, Lichtenstein E, Wilson GT. Understanding and preventing relapse. American Psychologist
1986;41(7):765-782. [doi: 10.1037//0003-066X.41.7.765]
XSL•FO
RenderX

30. Beck JS. The Complete Beck Diet for Life: the 5-stage program for permanent weight loss. Birmingham, AL: Oxmoor
House; 2008.
31. Ajzen I. The theory of planned behavior. Organizational Behavior and Human Decision Processes 1991 Dec;50(2):179-211.
[doi: 10.1016/0749-5978(91)90020-T]
32. Heshmat S. Eating Behavior and Obesity: Behavioral Economics Strategies for Health Professionals. New York: Springer
Publishing Company; 2013.
33. Baumeister RF, Tierney J. Chapter 10. The perfect storm of dieting. In: Willpower: Rediscovering the Greatest Human
Strength. New York: Penguin Books; 2011:214-237.
34. Cornum R, Matthews MD, Seligman ME. Comprehensive soldier fitness: building resilience in a challenging institutional
context. Am Psychol 2011 Jan;66(1):4-9. [doi: 10.1037/a0021420] [Medline: 21219042]
35. Seligman M, Railton P, Baumeister RF, Sripada C. Navigating Into the Future or Driven by the Past. Perspectives on
Psychological Science 2013 Feb 27;8(2):119-141. [doi: 10.1177/1745691612474317]
36. Kabat-Zinn J. Wherever You Go, There You Are: Mindfulness Meditation in Everyday Life. New York: Hyperion; 1994.
37. Herbert JD, Forman EM. Acceptance and Mindfulness in Cognitive Behavior Therapy: Understanding and Applying the
New Therapies. Hoboken, NJ: Wiley; 2011.
38. Glanz K, Bishop DB. The role of behavioral science theory in development and implementation of public health interventions.
Annu Rev Public Health 2010;31:399-418. [doi: 10.1146/annurev.publhealth.012809.103604] [Medline: 20070207]
39. Davis NJ, Tomuta N, Schechter C, Isasi CR, Segal-Isaacson CJ, Stein D, et al. Comparative study of the effects of a 1-year
dietary intervention of a low-carbohydrate diet versus a low-fat diet on weight and glycemic control in type 2 diabetes.
Diabetes Care 2009 Jul;32(7):1147-1152 [FREE Full text] [doi: 10.2337/dc08-2108] [Medline: 19366978]
Abbreviations
EHR: electronic health records
HDL: high-density lipoprotein
IVR: Interactive Voice Response
LDL: low-density lipoprotein
PAMF: Palo Alto Medical Foundation
PAMFRI: Palo Alto Medical Foundation Research Institute
RCT: randomized controlled trial
Edited by G Eysenbach; submitted 17.11.14; peer-reviewed by J Di Noia; accepted 25.11.14; published 21.01.15
Please cite as:
Block G, Azar KMJ, Block TJ, Romanelli RJ, Carpenter H, Hopkins D, Palaniappan L, Block CH
A Fully Automated Diabetes Prevention Program, Alive-PD: Program Design and Randomized Controlled Trial Protocol
JMIR Res Protoc 2015;4(1):e3
URL: http://www.researchprotocols.org/2015/1/e3/
doi:10.2196/resprot.4046
PMID:
©Gladys Block, Kristen MJ Azar, Torin J Block, Robert J Romanelli, Heather Carpenter, Donald Hopkins, Latha Palaniappan,
Clifford H Block. Originally published in JMIR Research Protocols (http://www.researchprotocols.org), 21.01.2015. This is an
open-access article distributed under the terms of the Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium,
provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information,
a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be
included.
XSL•FO
RenderX

Alive-PD protocol and descriptive paper

More Related Content

What's hot (20)

Viewers also liked (8)

Similar to Alive-PD protocol and descriptive paper (20)

Alive-PD protocol and descriptive paper