Are Conversational Agents Used at Scale by Companies Offering
Digital Health Services for the Management and Prevention of
Diabetes?
Roman Keller
1a
, Jiali Yao
1b
, Gisbert Wilhelm Teepe
2c
, Sven Hartmann
5d
,
Kim-Morgaine Lohse
2
, Florian von Wangenheim
1,2 e
, Falk Müller-Riemenschneider
1,3,4 f
,
Jacqueline Louise Mair
1,2,
*
g
and Tobias Kowatsch
1,2,5,
*
h
1
Future Health Technologies, Singapore-ETH Centre, Campus for Research Excellence and Technological Enterprise
(CREATE), Singapore
2
Center for Digital Health Interventions, Department of Management, Technology, and Economics, ETH Zurich,
Zurich, Switzerland
3
Saw Swee Hock School of Public Health National University of Singapore, Singapore
4
Yong Loo Lin School of Medicine, National University of Singapore, Singapore
5
Centre for Digital Health Interventions, Institute of Technology Management, University of St. Gallen,
St. Gallen, Switzerland
Keywords: Digital Health Companies, Healthcare, Type 2 Diabetes, Prevention, Management, Funding, Conversational
Agents.
Abstract: Successful interventions to prevent and manage type 2 diabetes rely on long-term, day-to-day decisions which
take place outside of clinical settings. In this context, human resources are difficult to scale up, and leveraging
Conversational agents (CAs) could be one way to scale up healthcare to tackle the emerging epidemic of type
2 diabetes. The objective of this paper is to assess the degree to which CAs are employed by top-funded digital
health companies that target the prevention and management of type 2 diabetes. Companies were identified
via two venture capital databases, i.e. Crunchbase Pro and Pitchbook. Two independent reviewers screened
results and the final list of companies was validated and revised by three independent digital health experts.
The companies’ digital services (usually mobile applications) were accessed and reviewed for the utilisation
of CAs. To better understand the purpose of identified CAs, relevant publications were identified via PubMed,
Google Scholar, ACM Digital Library and on the companies’ website. Nine out of 15 companies’ digital
services were accessible to the authors and only in one case a CA was employed. The uptake of CAs by top-
funded digital health companies targeting type-2 diabetes is still low.
1 INTRODUCTION
Diabetes is a chronic condition characterized by
elevated levels of blood glucose. It occurs when the
pancreas cannot produce sufficient hormone insulin
a
https://orcid.org/0000-0003-4810-4944
b
https://orcid.org/0000-0001-7269-4679
c
https://orcid.org/0000-0002-2264-9797
d
https://orcid.org/0000-0002-0256-6687
e
https://orcid.org/0000-0003-3964-2353
f
https://orcid.org/0000-0003-1402-7477
g
https://orcid.org/0000-0002-1466-8680
h
https://orcid.org/0000-0001-5939-4145
*
Shared last authorship
or the body cannot use the insulin effectively (WHO,
2016). If not managed appropriately, diabetes can
lead to a series of life-threatening microvascular and
macrovascular complications including blindness,
Keller, R., Yao, J., Teepe, G., Hartmann, S., Lohse, K., von Wangenheim, F., Müller-Riemenschneider, F., Mair, J. and Kowatsch, T.
Are Conversational Agents Used at Scale by Companies Offering Digital Health Services for the Management and Prevention of Diabetes?.
DOI: 10.5220/0010412708110816
In Proceedings of the 14th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2021) - Volume 5: HEALTHINF, pages 811-816
ISBN: 978-989-758-490-9
Copyright
c
2021 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
811
kidney failure, heart attacks, stroke and lower limb
amputation (WHO, 2016).
Type 2 diabetes (T2D) is the most common form
of diabetes and accounts for about 90% of diabetes
cases worldwide (International Diabetes Federation,
2019; WHO, 2016). Despite its links to non-
modifiable factors such as family history, ethnicity,
and increasing age, T2D is strongly tied to key
modifiable lifestyle risk factors including overweight
or obesity, poor diet, physical inactivity, and smoking
(International Diabetes Federation, 2019; WHO,
2016). Numerous landmark studies in the past have
demonstrated that over 50% of T2D can be prevented
or delayed through cornerstone interventions
targeting lifestyle modification and through
pharmacological interventions (International
Diabetes Federation, 2019). Likewise, strong
evidence also shows that T2D can be effectively
managed, or even reversed, in T2D patients via
similar intervention strategies (International Diabetes
Federation, 2019). Despite firm and encouraging
evidence, many key challenges persist with respect to
real-world implementation of T2D prevention and
management interventions at scale. One key
challenge is that conventional approaches of
delivering these T2D interventions are financially
expensive and human-resource intensive (Wu et al.,
2019). Another is that these interventions on T2D,
similar to those on other non-communicable diseases,
rely heavily on individual’s long-term and day-to-day
decisions, which mostly occur outside clinics and
hospitals and are difficult to monitor and intervene in
traditional manners (Rui et al., 2014; Wu et al., 2019).
Rapid advancements in digital and wireless
technology have provided unprecedented
opportunities to overcome such challenges and
transform and ‘scale-up’ healthcare delivery. This has
motivated large amounts of recent research and
investments of digital health companies into
developing digital solutions (Kvedar et al., 2016),
aiming to harness pervasive digital technology to
improve and scale up the prevention and management
of T2D (Alexander Fleming et al., 2020). Promising
results have been found in many digital programs of
T2D on glucose control, medication adherence,
weight loss, and quality of life (Alexander Fleming et
al., 2020). However, digital health programs have so
far been hampered by low user engagement and high
attrition rates, and long-term evidence is lacking to
comprehensively assess practical values (Garg &
Parkin, 2019). To address these issues, ‘virtual
diabetes coaching’ via Conversational Agents (CAs)
has been identified as an encouraging component of
current and future digital health offering for T2D care
(Garg & Parkin, 2019; O’brien, 2017; Ramchandani,
2019).
CAs are computer systems that imitate human
conversation using images, text or spoken language
and they can offer personalized human-like
interactions and social contacts (Laranjo et al., 2018;
Schachner et al., 2020). Recent studies have shown
promising results of CAs with regard to patient
satisfaction (T. W. Bickmore et al., 2010), treatment
outcomes (Ma et al., 2019), and the ability to build
work alliances with the patient (T. W. Bickmore &
Picard, 2005; Hauser-Ulrich et al., 2020; Kowatsch et
al., 2020). In addition, CAs have been found to be safe
for T2D care and even provide comparable
effectiveness to peer human coaching (Car et al.,
2020; King et al., 2020).
However, despite an increasing evidence base of
utilizing CAs for T2D prevention and management,
little is known about the actual adoption of CAs in the
burgeoning digital health industry and its most
successful companies. Moreover, there is a need to
identify the characteristics of those CAs in order to
guide future developments in the industry. Therefore,
this study aims to provide an observational analysis
of the uptake and characteristics of CAs among
digital services of the top-funded companies in the
T2D digital health industry.
2 METHODS
2.1 Databases and Companies
Top-funded digital health companies targeting the
prevention and management of T2D were identified
using the two venture capital databases Crunchbase
Pro and Pitchbook (Crunchbase, n.d.; PitchBook,
n.d.). These databases have been found to be among
the most comprehensive and accurate venture capital
databases and both are commonly used in academic
reports and by investors (Retterath & Braun, 2020).
We decided to select the fifteen companies with the
most funding in terms of total funding amount
through July 23, 2020.
2.1.1 Search Strategy
The search strategy included an extensive list of terms
describing the constructs “verticals & methods &
industries”, “diabetes”, and “management &
prevention”. The overview of the complete search
strategy for Crunchbase and Pitchbook is shown in
Table 1.
Scale-IT-up 2021 - Workshop on Scaling-Up Healthcare with Conversational Agents
812
Table 1: The search strategy used in Crunchbase and
Pitchbook.
Search
cate
g
or
y
Search terms
Verticals &
methods &
industries
Monitoring Equipment OR diagnostic
OR HealthTech OR healthcare
devices OR connected health* OR
Therapeutic Devices OR Digital
Health OR digital health* OR health*
technology OR health* app* OR
wearables OR Mobile health OR
mhealth OR mobile app OR personal
health OR virtual care OR e-health
OR assistive technology OR
telehealth OR telemedicine OR
health* platform OR healthcare it OR
data management OR Artificial
Intelligence & Machine Learning OR
Cloud data services OR analytics OR
health* diagnostics OR Big Data OR
information OR digital OR data OR
biometrics OR home health care OR
medtech OR self-monitorin
g
Diabetes obesity OR blood sugar OR blood
g
lucose OR insulin OR diabet*
Management
&
Prevention
diabetes management OR diabetes
treatment OR diabetes control OR
diabetes monitoring OR blood sugar
monitoring OR disease monitoring
OR disease management OR risk
reduction OR disease prevention OR
diabetes prevention OR prevention
OR prediabet*
2.1.2 Selection Criteria
We included companies if they offered a digital
health intervention that involved the prevention or
management of T2D. A digital health intervention
was defined as a discrete functionality of digital
technology that is applied to achieve health objectives
and followed the definition of the WHO classification
of digital health interventions (World Health
Organization, 2018).
Companies were excluded if their intervention (1)
did not focus on patients; (2) mainly focused on the
administrational needs of hospitals (hospital
information systems); and (3) did not involve a digital
solution as main intervention component. In addition,
we excluded companies if their intervention (4)
involved any form of continuous blood glucose
monitoring or automated insulin delivery system; (5)
did not involve behaviour change; and (6) mainly
targeted insulin-dependent diabetics, as such
interventions mainly target medication adherence
rather than lifestyle behaviours.
Furthermore, we excluded companies in which
the intervention was any kind of general weight loss
or fitness app without focus on T2D.
2.1.3 Selection Process
The top 15 most-funded companies combined from
both database searches were identified as they
account for over 90% of the total funding in all
companies which gives a representative and
comprehensive insight into current state of the digital
health industry. In case of conflicting funding
information between both databases, Crunchbase
information was preferred as it has been found to have
better coverage with respect to total capital
committed and financing rounds than Pitchbook
(Retterath & Braun, 2020). Eligibility of companies
was determined by two independent investigators.
The results were compared, and disagreements
discussed until a consensus was reached.
After a first list was defined, three independent
experts in the fields of digital health and T2D were
approached to validate the company list. In case of
disagreements between experts, the authors followed
the opinions of the majority. Based on their feedback,
the company selection criteria were slightly adapted,
and the company list was adapted. In particular,
companies that provide interventions for continuous
glucose measurement and automated insulin delivery
systems were excluded from our list, as their digital
solution was not considered to be the main
intervention component. In addition, the services of
those companies mainly focused on insulin-
dependent diabetics and were not involving any
behavioural change strategies. Furthermore, the
company KKT Technology Pte. Ltd. (Holmusk) was
added to the list of top-funded digital health
companies, as the company meets our inclusion
criteria in terms of the amount of funding and offered
digital health intervention but was not identified with
our search strategy.
2.2 Digital Health Interventions
All identified digital services were available in the
form of mobile applications as predominant mode of
intervention delivery. Thus, we searched and
downloaded these apps from the two most popular
app stores: Google Play store and Apple app store
(IDC, 2020). The latest app search was conducted on
November 21, 2020.
If a mobile application was not accessible to the
authors, the companies were systematically
approached via email and requested for access. In
Are Conversational Agents Used at Scale by Companies Offering Digital Health Services for the Management and Prevention of Diabetes?
813
case no reply was received on the first email, a
follow-up email was sent two weeks later.
If the authors were not able to access the app, no
judgement on the presence of a CA could be done.
2.3 Publications
Three databases (PubMed, Google Scholar, and ACM
Digital Library) were systematically searched for
scientific articles, published between database
inception and October 2020, using the search terms
“Name_Intervention” AND (Smartphone OR
Application OR App OR Intervention OR Mobile
Health) relating to the identified company’s T2D app.
In addition, we screened the websites of the
companies for relevant publications on their
conversational agents, as some publications could not
be identified via the databases searches.
2.4 Conversational Agents
Two investigators reviewed the identified accessible
apps and assessed whether a conversational agent was
utilised (Table 2). In a second step, the characteristics
of the identified CA were extracted by reviewing the
app as well as the identified publications. The
conversational agent characteristics included the
health goal, type of agent, input format, output format
and dialogue initiative. Covidence
(www.covidence.org) was used for the data
extraction.
3 RESULTS
The overview of the 15 top-funded digital health
companies in the prevention and management of T2D
and their utilised CAs can be seen in Table 2.
From the reviewed companies, 6 apps were not
accessible because they were only available: (1) with
a subscription service (Virta Health Corp., Dario
Health Corp.); (2) in a specific geographic region
(Virta Health Corp., Welldoc Inc., Liva Healthcare
ApS); or (3) with an employer subscription or when
being referred by a physician (Virta Health Corp.,
Dario Health Corp., Welldoc Inc., Twin Health Inc.,
Sweetch Health Ltd.). Of the 9 accessible apps, 8 apps
did not include any CAs, and Lark Technologies, Inc.
was the only company with an app that employed a
conversational agent.
As Lark Technologies, Inc. was the only company
with a diabetes-relevant conversational agent, only
publications of Lark Technologies, Inc. were
considered for this analysis.
Table 2: Ranking of the fifteen top-funded digital health
companies in type 2 diabetes prevention and management
and utilisation of CAs in the mobile applications of the
companies as of November 27, 2020, sorted in descending
order of total funding amount. CA = Conversational agent,
Y = CA used, N = No CA used, - = Mobile application not
accessible.
Company
Funding
Rank #
Company legal name
CA
used?
1 Omada Health, Inc. N
2 Livon
g
o Health, Inc. N
3 Virta Health Corp. -
4 Noom, Inc. N
5 DarioHealth Cor
p
. -
6 Informed Data Systems, Inc.
(One Drop)
N
7 Lark Technologies, Inc. Y
8 Vida Health, Inc. N
9 Welldoc, Inc. -
10 Twin Health, Inc. -
11 Oviva, Inc. N
12 KKT Technology Pte. Ltd.
(
Holmusk
)
N
13 Fruit Street Health P.B.C. N
14 Sweetch Health Ltd. -
15 Liva Healthcare ApS -
Lark Technologies, Inc. offers a digital care platform
for chronic conditions and has specialised health
plans for patients with diabetes, hypertension,
prediabetes and general health risk behaviours such
as stress and anxiety, smoking cessation or weight
management. Lark is accessible to all users via their
employer, health plan, provider or individually
through a subscription. A basic version of the app is
free and available to everyone.
Two publications of Lark Technologies, Inc. were
identified via the company website and both papers
were not published in peer-reviewed journals (Stein
et al., 2019, 2020).
The goals of Lark’s conversational agent
comprised assistance, training, education, prevention
and onboarding. It utilized a counselling type of agent
with a mixed dialogue initiative (system and user
were both able to start a conversation). The input
format allowed for fixed text with predefined answer
options and external input from sensors (E.g.
smartphone accelerometer, location sensor). The
output format was only in text.
Scale-IT-up 2021 - Workshop on Scaling-Up Healthcare with Conversational Agents
814
4 DISCUSSION
To our knowledge, this is the first study to identify
the degree to which the most successful companies
offering digital health interventions for the prevention
and management of T2D employ CAs. Our study
found that out of nine accessible diabetes apps, only
in one case, a CA is used. This indicates that the T2D
industry uptake of CAs still remains low.
This finding corroborates recent research findings of
literature reviews investigating the state of research
on CAs in healthcare which is still considered to be in
an early developing stage (Bérubé et al., 2020; Car et
al., 2020; Laranjo et al., 2018; Montenegro et al.,
2019; Schachner et al., 2020). With specific focus on
T2D, most reviews only identified a small number of
CAs targeting T2D. For example, a scoping review
by Car et al. only identified two studies involving a
conversational agent to target T2D prevention or
management (Car et al., 2020). Interestingly, one of
the identified studies by Car et al. (Car et al., 2020)
assessed a digital health intervention from Wellthy
(Wellthy Therapeutics Private Limited) which would
have been a relevant company for our study (Sosale
et al., 2018). However, due to insufficient funding
Wellthy did not appear on our list of the top-funded
digital health companies targeting T2D. This shows
that our approach to only focus on the top funded
companies may have left out certain relevant
companies that employ CAs for T2D management or
prevention.
Nonetheless, the general lack of CAs in the identified
digital health interventions is remarkable. One
potential reason for the low utilisation of CAs could
be safety concerns for patients and consumers when
relying on actionable medical information from CAs
(T. W. Bickmore et al., 2018).
The low utilisation of CAs could limit the
interventions’ scalability as traditional nondigital
approaches are generally time and resource intensive.
For example, the use of human health experts is
expensive, and it requires considerable time resources
from experts to provide personalised lifestyle
coaching to each intervention user. In addition, CAs
have recently shown promising results related to
patient acceptance (T. W. Bickmore et al., 2010) and
on building work alliances with patients (T. Bickmore
et al., 2005). Therefore, we advise digital health
companies to increasingly consider CAs in their
digital health interventions for the prevention and
management of T2D.
5 CONCLUSIONS
The use of CAs in digital health interventions of top-
funded digital health companies targeting the
prevention and management of T2D still remains low.
Digital Health companies should increasingly
consider the use of CAs in their interventions to
increase scalability by reducing costs and time
resources.
ACKNOWLEDGEMENTS
This research is supported by the National Research
Foundation, Prime Minister’s Office, Singapore
under its Campus for Research Excellence and
Technological Enterprise (CREATE) programme.
REFERENCES
Alexander Fleming, G., Petrie, J. R., Bergenstal, R. M.,
Holl, R. W., Peters, A. L., & Heinemann, L. (2020).
Diabetes digital app technology: Benefits, challenges,
and recommendations. A consensus report by the
European Association for the Study of Diabetes
(EASD) and the American Diabetes Association
(ADA) Diabetes Technology Working Group. Diabetes
Care, 43(1), 250–260. https://doi.org/10.2337/dci19-
0062
Bérubé, C., Schachner, T., Keller, R., Fleisch, E., von
Wangenheim, F., Barata, F., & Kowatsch, T. (2020).
Voice-based Conversational Agents for the Prevention
and Management of Chronic and Mental Conditions: A
Systematic Literature Review. JMIR Preprints
30/11/2020:25933.
https://doi.org/10.2196/preprints.25933
Bickmore, T., Gruber, A., & Picard, R. (2005). Establishing
the computer-patient working alliance in automated
health behavior change interventions. Patient
Education and Counseling, 59(1), 21–30.
https://doi.org/10.1016/j.pec.2004.09.008
Bickmore, T. W., Mitchell, S. E., Jack, B. W., Paasche-
Orlow, M. K., Pfeifer, L. M., & O’Donnell, J. (2010).
Response to a relational agent by hospital patients with
depressive symptoms. Interacting with Computers,
22(4), 289–298.
https://doi.org/10.1016/j.intcom.2009.12.001
Bickmore, T. W., & Picard, R. W. (2005). Establishing and
maintaining long-term human-computer relationships.
ACM Transactions on Computer-Human Interaction,
12(2), 293–327.
https://doi.org/10.1145/1067860.1067867
Bickmore, T. W., Trinh, H., Olafsson, S., O’Leary, T. K.,
Asadi, R., Rickles, N. M., & Cruz, R. (2018). Patient
and consumer safety risks when using conversational
assistants for medical information: An observational
Are Conversational Agents Used at Scale by Companies Offering Digital Health Services for the Management and Prevention of Diabetes?
815
study of siri, alexa, and google assistant. Journal of
Medical Internet Research, 20(9), e11510.
https://doi.org/10.2196/11510
Car, L. T., Dhinagaran, D. A., Kyaw, B. M., Kowatsch, T.,
Joty, S., Theng, Y. L., & Atun, R. (2020).
Conversational agents in health care: Scoping review
and conceptual analysis. In Journal of Medical Internet
Research (Vol. 22, Issue 8, p. e17158). JMIR
Publications. https://doi.org/10.2196/17158
Crunchbase. (n.d.). Crunchbase [home page on the
internet]. Retrieved November 29, 2020, from
https://www.crunchbase.com/
Garg, S. K., & Parkin, C. G. (2019). The Emerging Role of
Telemedicine and Mobile Health Technologies in
Improving Diabetes Care. Diabetes Technology and
Therapeutics, 21(S2), S2-1-S2-3.
https://doi.org/10.1089/dia.2019.0090
Hauser-Ulrich, S., Künzli, H., Meier-Peterhans, D., &
Kowatsch, T. (2020). A smartphone-based health care
chatbot to promote self-management of chronic pain
(SELMA): Pilot randomized controlled trial. JMIR
MHealth and UHealth, 8(4), e15806.
https://doi.org/10.2196/15806
IDC. (2020). Introduction - WHO guideline
Recommendations on Digital Interventions for Health
System Strengthening - NCBI Bookshelf.
https://www.ncbi.nlm.nih.gov/books/NBK541905/
International Diabetes Federation. (2019). IDF Diabetes
Atlas 9th edition.
King, A. C., Campero, M. I., Sheats, J. L., Castro Sweet, C.
M., Hauser, M. E., Garcia, D., Chazaro, A., Blanco, G.,
Banda, J., Ahn, D. K., Fernandez, J., & Bickmore, T.
(2020). Effects of counseling by peer human advisors
vs computers to increase walking in underserved
populations: The COMPASS randomized clinical trial.
JAMA Internal Medicine, 180(11), 1481–1490.
https://doi.org/10.1001/jamainternmed.2020.4143
Kowatsch, T., Schachner, T., Harperink, S., Barata, F.,
Dittler, U., Xiao, G., Stanger, C., von Wangenheim, F.,
Fleisch, E., Oswald, H., & Möller, A. (2020).
Conversational Agents as Mediating Social Actors in
Chronic Disease Management Involving Healthcare
Professionals, Patients, and Family Members. JMIR
Preprints #25060.
https://preprints.jmir.org/preprint/25060
Kvedar, J. C., Fogel, A. L., Elenko, E., & Zohar, D. (2016).
Digital medicine’s march on chronic disease. Nature
Biotechnology, 34(3), 239–246.
https://doi.org/10.1038/nbt.3495
Laranjo, L., Dunn, A. G., Tong, H. L., Kocaballi, A. B.,
Chen, J., Bashir, R., Surian, D., Gallego, B., Magrabi,
F., Lau, A. Y. S., & Coiera, E. (2018). Conversational
agents in healthcare: A systematic review. In Journal of
the American Medical Informatics Association (Vol.
25, Issue 9, pp. 1248–1258). Oxford University Press.
https://doi.org/10.1093/jamia/ocy072
Ma, T., Chattopadhyay, D., & Sharifi, H. (2019). Virtual
humans in health-related interventions: A meta-
analysis. Conference on Human Factors in Computing
Systems - Proceedings, 1–6.
https://doi.org/10.1145/3290607.3312853
Montenegro, J. L. Z., da Costa, C. A., & da Rosa Righi, R.
(2019). Survey of conversational agents in health. In
Expert Systems with Applications (Vol. 129, pp. 56–
67). Elsevier Ltd.
https://doi.org/10.1016/j.eswa.2019.03.054
O’brien, J. D. (2017). Chatbots for Diabetes Self-
Management: Diabetes coaching at scale.
PitchBook. (n.d.). PitchBook [home page on the Internet].
Retrieved November 29, 2020, from
https://pitchbook.com/
Ramchandani, N. (2019). Virtual Coaching to Enhance
Diabetes Care. Diabetes Technology and Therapeutics,
21(S2), S2-48-S2-51.
https://doi.org/10.1089/dia.2019.0016
Retterath, A., & Braun, R. (2020). Benchmarking Venture
Capital Databases. https://ssrn.com/abstract=3706108
Rui, P., Hing, E., & Okeyode, T. (2014). National
Ambulatory Medical Care Survey: 2014 State and
National Summary Tables.
https://www.cdc.gov/nchs/ahcd/ahcd_products.htm
Schachner, T., Keller, R., & V Wangenheim, F. (2020).
Artificial Intelligence-Based Conversational Agents for
Chronic Conditions: Systematic Literature Review.
Journal of Medical Internet Research, 22(9), e20701.
https://doi.org/10.2196/20701
Sosale, A. R., Shaik, M., Shah, A., Chawla, R., Makkar, B.
M., Kesavadev, J., Joshi, S., Deshpande, N., Agarwal,
S., Mahseshwari, A., Madhu, S., & Saboo, B. D. (2018).
Real-World Effectiveness of a Digital Therapeutic in
Improving Glycaemic Control in South Asians Living
with Type 2 Diabetes. Diabetes, 67(Supplement 1),
866-P. https://doi.org/10.2337/db18-866-p
Stein, N., Delury, K., & Paruthi, J. (2020). One-Year
Clinical Outcomes of an Artificial Intelligence-Based
Digital Diabetes Prevention Program.
Stein, N., Ku, R., & Mao, T. (2019). Clinical outcomes from
older adults in a digital diabetes pre-vention program.
https://doi.org/10.2337/dci18-0007
WHO. (2016). WHO Global Report. Global Report on
Diabetes.
World Health Organization. (2018). Classification of
Digital Health Interventions v 1.0: a shared language
to describe the uses of digital technology for health.
http://apps.who.int/iris/bitstream/handle
/10665/260480/WHO-RHR-18.06-eng.pdf
Wu, X., Guo, X., & Zhang, Z. (2019). The efficacy of
mobile phone apps for lifestyle modification in
diabetes: Systematic review and meta-analysis. In JMIR
mHealth and uHealth (Vol. 7, Issue 1). JMIR
Publications. https://doi.org/10.2196/12297
Scale-IT-up 2021 - Workshop on Scaling-Up Healthcare with Conversational Agents
816