Abstract
Introduction
This article is our annual attempt to highlight the key papers written between July 1, 2019, and June 30, 2020, in which digital health technologies were used to prevent or treat diabetes mellitus. This article will mainly address interventions called digital therapeutics, while other articles will address telemedicine (remote patient–provider visits) and approaches to automated or assisted insulin delivery approaches.
Digital therapeutics (DTx) deliver evidence-based therapeutic interventions to patients that are driven by high quality software programs to prevent, manage, or treat a medical disorder or disease. They are used independently or together with medications, devices, or other therapies to optimize patient care and health outcomes. (DTx Alliance;
The best digital therapeutics have specific mechanisms of action based on theories of health behavior with documented outcomes measured in ways analogous to those used for pharmaceuticals or devices. The field, which is about 5 years old, has had very good and occasionally tremendous clinical and business outcomes. However, the research documenting many of the outcomes and claims is early in its development. Models for reimbursement by insurance companies and government programs are also immature and emerging. Research documenting outcomes and advocacy for proven-effective approaches are needed at multiple levels.
This year's selection of potential articles to include in this article was limited by the effects of the COVID-19 pandemic. The incredible number of papers about meeting the acute medical needs (inpatient and outpatient) of people with diabetes was breathtaking. However, the number of articles about digital therapeutics (typically addressing longer-term treatments and behavior change) fell precipitously as researchers reported on the critical needs to help save lives during the pandemic.
Fleming GA, Petrie JR, Bergenstal RM, Holl RW, Peters AL, Heinemann L
Levine BJ, Close KL, Gabbay RA
Mayberry LS, Lyles CR, Oldenburg B, Osborn CY, Parks M, Peek ME
Fagherazzi G, Ravaud P
Celik A, Forde R, Sturt J
Buysse H, Coremans P, Pouwer F, Ruige J
Gimbel RW, Rennert LM, Crawford P, Little JR, Truong K, Williams JE, Griffin SF, Shi L, Chen L, Zhang L, Moss JB, Marshall RC, Edwards KW, Crawford KJ, Hing M, Schmeltz A, Lumsden B, Ashby M, Haas E, Palazzo K
Sartori AC, Rodrigues Lucena TF, Takáo Lopes C, Picinin Bernuci M, Yamaguchi MU
Kassavou A, Mirzaei V, Brimicombe J, Edwards S, Massou E, Prevost AT, Griffin S, Sutton S
Khalil C
Lemelin A, Paré G, Bernard S, Godbout A
Sunil Kumar D, Prakash B, Subhash Chandra BJ, Kadkol PS, Arun V, Thomas JJ
Kim SH, Utz S
Reidy C, Klonoff DC, Barnard-Kelly KD
Hammersley ML, Okely AD, Batterham MJ, Jones RA
Salvy SJ, Carandang K, Vigen CL, Concha-Chavez A, Sequeira PA, Blanchard J, Diaz J, Raymond J, Pyatak EA
Forman EM, Goldstein SP, Crochiere RJ, Butryn ML, Juarascio AS, Zhang F, Foster GD
Dunn CG, Turner-McGrievy GM, Wilcox S, Hutto B
McMullan M, Millar R, Woodside JV
Rewolinski JA, Kelemen A, Liang Y
Gao Z, Pope ZC, Lee JE, Quan M
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
Fleming GA1, Petrie JR2, Bergenstal RM3, Holl RW4, Peters AL5, Heinemann L6
1Kinexum, Harpers Ferry, WV; 2Institute of Cardiovascular and Medical Sciences, University of Glasgow, Glasgow, UK; 3International Diabetes Center at Park Nicollet, Minneapolis, MN; 4Institute of Epidemiology and Medical Biometry, ZIBMT, University of Ulm, Ulm, Germany; 5Keck School of Medicine of the University of Southern California, Los Angeles, CA; 6Science-Consulting in Diabetes GmbH, Neuss, Germany
Background
To help people manage their diabetes, digital and health applications (“apps”) are quickly being developed. Many health-related apps offered on smartphones and other wireless devices are available to support people with diabetes who need to adopt lifestyle interventions or medication adjustments in response to glucose-monitoring data. However, regulations and guidelines have yet to standardize how mobile health apps are assessed for patient safety and clinical validity. The available evidence on the safety and effectiveness of mobile health apps, especially for diabetes, remains limited.
Methods
The European Association for the Study of Diabetes (EASD) and the American Diabetes Association (ADA) have performed a joint review of the current landscape of available diabetes digital health technology (which included only stand-alone diabetes apps, as opposed to those that are integral to a regulated medical device, such as insulin pumps, continuous glucose monitoring systems, and automated insulin delivery systems) and practices of regulatory authorities and organizations.
It seems that across the United States and Europe, mobile apps for managing health and wellness are mostly unregulated unless they meet the definition of medical devices for therapeutic and/or diagnostic purposes. International organizations—including the International Medical Device Regulators Forum and the World Health Organization—have made advances in classifying various kinds of digital health technology and assimilating digital health technology into the field of medical devices.
Results
As the diabetes digital health field grows and becomes more fully integrated into daily life, we hope to make sure that it is based on the best evidence for safety and efficacy. Therefore, we present several concerns that the diabetes community, including regulatory authorities, policy-makers, professional organizations, researchers, people with diabetes, and healthcare professionals, should address to ensure that diabetes health technology can meet its full potential. These issues range from inadequate evidence on app accuracy and clinical validity to lack of training provision, poor interoperability and standardization, and insufficient data security.
Conclusion
A series of recommended actions to resolve some of these shortcomings conclude this study.
Comment
This consensus report highlights an important issue: digital health technologies, especially mobile health (mHealth) applications for smartphones, are proliferating at an astounding rate, yet the evidence to support their use by persons with diabetes or their clinical teams lags painfully behind. This creates a conundrum: a significant fraction of the healthcare economy involves investment in these new technologies without a proportional increase in the investment of financial resources to rigorously study their safety and efficacy. There is an urgent need for regulatory bodies to improve the classification of mHealth applications, especially those that integrate with medical devices or data derived from medical devices in some way. One should pay particular attention to the article's emphasis on the need for app developers to also develop proper training toolkits, and on the need for medical device manufacturers and mHealth app developers to focus on data interoperability and standardization. I find only one item missing from the recommendation: that regulatory authorities should compel device manufacturers to allow persons with diabetes to share their data with other digital devices and mHealth apps. After all, the person with diabetes should be the final arbiter and authority over the use of data that were generated from their bodies (or minds, in the case of patient-reported outcomes).
Reviewing U.S. connected diabetes care: the newest member of the team
Levine BJ1, Close KL1, Gabbay RA2
1Close Concerns, San Francisco, CA; 2Joslin Diabetes Medical Center, Harvard Medical School, Boston, MA
This manuscript is also discussed in the article on Primary Care and Diabetes Technologies and Treatments, page S-143.
Background
Recent years have brought about an explosion in the number of companies offering connected diabetes care products, defined as digital diabetes management systems, based around (1) smartphone apps, (2) devices with built-in connectivity, and (3) remote human and automated coaching and support. These nascent models aim to provide more continuous and on-demand care, aligning with the 24/7 demands of chronic disease. It has been enabled by multiple factors, including the rising use of connected devices and apps to help people manage their chronic conditions, growing appreciation for the importance of outcomes beyond A1c, and the lofty and growing cost of healthcare. Despite the potential of these programs to improve the outcomes and well-being of people with diabetes and reduce the burdens on healthcare providers and systems, awareness and use of these programs and approaches remain low in the medical community.
Methods
In this article, we present a snapshot of this dynamic field, including a taxonomy of various connected diabetes care products available to employers, health plans, health systems, and people with diabetes in the United States.
Results
This study identifies the following meaningful distinctions in this field: (1) health conditions managed, (2) peer support interactions, (3) prescribing providers on the care team, (4) provision of connected medical devices and/or continuous glucose monitors, (5) degree of treatment personalization, and (6) clinical and real-world evidence. In addition, we discuss broad trends in connected diabetes care.
Conclusion
Given the urgency and scale of the diabetes epidemic, it is vital that a range of medical and clinical communities find meaningful ways to scale individualized, timely care under reimbursement models that better align incentives for various stakeholders, particularly healthcare providers themselves. This would not only address deficiencies in care but could also make diabetes care more attractive to future clinicians.
Comment
The authors present a concise overview of the state of digital interventions designed to prevent or treat diabetes. They conclude there are a number of factors enabling remote connected care including: (1) more connected devices, (2) patient demand, (3) cost of diabetes care, (4) outcome measures beyond A1C, (5) range of approaches available, (6) shortage of providers, and (7) investment in digital health and therapeutics. They highlight the importance of peer support as major elements of many interventions. Another trend to watch is the way digital health companies are expanding their reach by either providing more of the clinical care a patient needs and/or by addressing the needs of patients with a variety of chronic diseases beyond diabetes. These trends may become accelerated as well-funded start-up companies acquire new services and/or are acquired by more traditional healthcare delivery approaches. Regardless, we are in for a wild ride in which the road is being built while also being driven on by entrepreneurs and established business. Hoping no one crashes.
mHealth interventions for disadvantaged and vulnerable people with type 2 diabetes
Mayberry LS1, Lyles CR2, Oldenburg B3, Osborn CY4, Parks M1, Peek ME5
1Vanderbilt University Medical Center, Nashville, TN; 2University of California, San Francisco, CA; 3The University of Melbourne, Victoria, Australia; 4Lirio, Nashville, TN; 5Section of General Internal Medicine, Chicago Center for Diabetes Translation Research, The University of Chicago, Chicago, IL
Background
Mobile- and Internet-delivered (collectively, digital) interventions are commonly used by persons with diabetes (PWD) to help self-manage and improve/maintain glycemic control (hemoglobin A1c [HbA1c]). However, evidence of the acceptance and benefits of such interventions among disadvantaged/vulnerable PWD is still quite limited.
Methods
We reviewed studies published from 2011–April 2019 that assessed the impact of diabetes self-management interventions provided via mobile device and/or the Internet on the glycemic control of disadvantaged/vulnerable adults with type 2 diabetes (T2D). The studies that were included reported ≥50% of the sample having a low socioeconomic status and/or being a racial/ethnic minority, or living in a rural setting or low-/middle-income country (LMIC). We identified 21 studies evaluating a digital intervention among disadvantaged/vulnerable PWD.
Results
Although many digital interventions found A1c improvements within groups (16 of 21 studies), only 7 of the 17 studies with a control group found differences in A1c between groups. Three studies found reductions in emergency room (ER) visits and hospitalizations. We integrate this information and make recommendations for increasing access and improving the design and usability of such interventions. We also review the role of human support in digital delivery and issues related to study design, reporting, economic value, and available research in LMICs.
Conclusion
Evidence shows that digital interventions can improve diabetes control, healthcare utilization, and healthcare costs. More research is needed to support these early findings, and many issues remain concerning optimizing the impact of digital interventions on the health outcomes of disadvantaged/vulnerable persons with diabetes.
Comment
This article reviews the literature regarding mHealth interventions in disadvantaged and vulnerable populations, including those living in low- to middle-income countries. Not surprisingly, the studies reviewed demonstrated mixed results given the diverse nature of their approaches. Importantly, the authors lay out some guiding principles that could lead to increased availability and effectiveness of digital intervention.
Technology can be used to engage populations who have been historically difficult to reach via traditional healthcare delivery or communication channels.
Technology can automate the delivery of health information and support, with relatively limited demand on healthcare system resources.
The rapid adoption of mobile devices throughout the world is redefining how healthcare is delivered and how people with diabetes manage their disease.
There are a variety of barriers for disadvantaged and vulnerable populations to be able to access online resources—not the least of which is lack of access to broadband Internet. This must be corrected as if our lives—or at least the lives of those most in need—depended on it.
Digital diabetes: perspectives for diabetes prevention, management and research
Fagherazzi G1, Ravaud P2
1Inserm UMR1018, Paris South, Paris Saclay University, Centre of Research in Epidemiology and Population Health (CESP), Villejuif, France; 2Inserm UMR1153, Paris Descartes University, AP–HP, Centre of Research in Epidemiology and Statistics, Sorbonne Paris Cité (CRESS), Paris, France
Background
Modern technology has the ability to transform the field of diabetes by implementing continuous and no-burden remote monitoring of patients' symptoms, physiological data, behaviors, and social and environmental contexts with wearables, sensors, and smartphone tools. Additionally, data procured online and via digital technologies, which the authors suggest be grouped under the term “digitosome,” constitute, through the quantity and variety of information they represent, great potential for identifying new digital markers and patterns of risk that can improve diabetes management and quality of life, and ultimately prevent diabetes-related complications.
Methods
Moving from a world in which patients are characterized by only a few recent measurements of fasting glucose levels and glycated hemoglobin to a world where we can consider various key parameters at thousands of time points simultaneously will profoundly change the way diabetes is prevented, managed, and characterized in patients living with diabetes, as well as how it is scientifically researched.
Results
This review considers how the digitization of diabetes can impact all aspects of diabetes—its prevention, management, technology, and research—and how it can complement, but not replace, traditional clinical care.
Conclusion
This might be a genuine game changer that should be embraced by all, as it can deliver solid research results transferable to patients, improve general health literacy, and provide tools to assist the everyday decision-making process by both healthcare professionals and patients living with diabetes.
Comment
This article is a thought/perspective piece. These authors review various digital technologies, their domains of impact, and their achievements in diabetes care to date. They cover obvious technological advances like continuous/flash glucose monitoring and closed-loop insulin delivery systems, but also more emerging technologies and their applications, such as:
AI algorithms that can predict glycemic responses to foods, automatically categorize retinal images, and provide optimized decision support to clinicians
Social media and online communities as sources of peer support
The authors propose the term “digitosome,” which represents all the data generated by digital technologies. This adds to the already crowded field of “omics”-related terms, including the well-known terms genome, transcriptome, metabolome, and proteome, but also emerging terms like behaviorome, envirome, and chronome (a term representing the longitudinal capture of any human-generated signal). When the big data companies learn to capture complete audio or video recordings of our clinical encounters, I anticipate that clinicians won't need to transcribe or document the medical record anymore. But I also anticipate that we will be adding new vocabulary words to our lectionary, such as “conversationome” and the “videographome.”
The impact of online self-management interventions on midlife adults with type 2 diabetes: a systematic review
Celik A, Forde R, Sturt J
Florence Nightingale Faculty of Nursing, Midwifery and Palliative Care, King's College London, UK
Background
Self-management programs online are recommended for people with type 2 diabetes to improve self-management capability, yet there is little proof that such programs improve the health outcomes for midlife adults with diabetes. The purpose of this systematic review was to determine the impact of online self-management interventions with digital consulting on HbA1c, total cholesterol, blood pressure, diabetes distress, self-efficacy, and depression in midlife adults.
Methods
This study was a systematic review that involved searching Medline, Embase, and CINAHL. The studies were evaluated using the Cochrane collaboration tool.
Results
In the eight studies that were included, online interventions led to the improvement of HbA1c (pooled mean difference on HbA1c: −0.35%, 95% CI (−0.52, −0.18), P<0.001). A narrative synthesis was conducted for all secondary outcomes. No conclusions could be drawn on the impact of these outcomes.
Conclusion
Online interventions improve HbA1c. Secondary outcomes require further research.
Comment
This review analyzed eight randomized control trials (RCTs) to assess the impact of online self-management interventions with digital consulting on HbA1c, total cholesterol, blood pressure, diabetes distress, self-efficacy, and depression in midlife adults. The review was not able to demonstrate clinically significant improvements, which is not surprising given the limited number of studies and the complexity of interpreting studies with different approaches and outcome measurements. The paper does summarize key issues seen in online interventions and how to measure outcomes that matter beyond glucose control. I realize that RCTs are the gold standard; however, they are quite difficult and expensive to do especially when a digital therapeutic is being evaluated. By the time a study is planned, funded, completed, and published, the technology and approaches used in the intervention are probably two to three generations beyond what are currently being used. Other research approaches such as propensity-matched comparison groups should be considered as worthy of inclusion is a systematic review.
Sustainable improvement of HbA1c and satisfaction with diabetes care after adding telemedicine in patients on adaptable insulin regimens: results of the TeleDiabetes randomized controlled trial
Buysse H1, Coremans P2, Pouwer F3, Ruige J2
1Ghent University, Ghent, Belgium; 2AZ Nikolaas, Sint-Niklaas, Belgium; 3University of Southern Denmark, Odense, Denmark
Background
This was a 2-year study to determine whether tele-education improves and helps maintain good glycemic control and patient satisfaction.
Methods
Randomly assigned adult patients either obtained immediate access to tele-education or were provided surplus education after 3 months (the control group). Clinical data were retrieved and patients completed questionnaires. Researchers conducted multivariate analyses of covariance and repeated measures analysis of variance.
Results
Implementation of tele-education in-between face-to-face contacts improved glycemic control for both groups, which was maintained over a 2-year period. Tele-education did not have an effect on glucose measurements or on hypoglycemic events. Patients were satisfied with this tele-educational tool and were pleased with the use of personal messages.
Conclusions
Additional research should look at the potential influence of “life changes” and influence on the “need for more tele-educational feedback,” and therefore on the provision of (mobile) platforms adaptable to patient's (changing life) situations.
Comment
This study addresses a question we have all asked ourselves: Do tele-education encounters in-between face-to-face clinical encounters improve glycemic control? What about patient satisfaction? Tele-education contacts, which typically include data sharing and some form of communication, can theoretically be initiated by the person with diabetes or by the care team. The present study randomized 153 Dutch-speaking individuals with type 1 or type 2 diabetes to treatment (patient-initiated tele-education contacts with data sharing) or control (current standard of care with in-clinic encounters only). The authors wisely decided to design the trial as a waitlist control, with the control group crossing over to the intervention after 3 months. The authors then monitored outcomes for 2 years from baseline and showed that the intervention was associated with a significant and sustained improvement in HbA1c, with A1c declining from approximately 62 to 55 mmol/mol (7.8% to 7.2%) over the 2-year period and most of the improvement occurring within the first 3–9 months after intervention exposure. It seems to me that diabetes teams spend a lot of time providing phone, video, or message-based tele-education. Now we have some emerging evidence to support going the extra mile for our patients.
Enhancing patient activation and self-management activities in patients with type 2 diabetes using the US Department of Defense mobile health care environment: feasibility study
Gimbel RW1, Rennert LM1, Crawford P2, Little JR3, Truong K1, Williams JE1, Griffin SF1, Shi L1, Chen L4, Zhang L5, Moss JB2, Marshall RC6, Edwards KW1, Crawford KJ2, Hing M7, Schmeltz A3, Lumsden B1, Ashby M1, Haas E1, Palazzo K1
1Department of Public Health Sciences, Clemson University, Clemson, SC; 2Nellis Family Medicine Residency Program, Mike O'Callaghan Federal Hospital, Las Vegas, NV; 3Mobile Health Innovation Center, Telemedicine & Advanced Technologies Research Center, U.S. Army Medical Research & Materials Command, Fort Gordon, GA; 4Fielding School of Public Health, University of California Los Angeles, Los Angeles, CA; 5College of Nursing and Health Sciences, University of Massachusetts Boston, Boston, MA; 6Clinical Informatics Fellowship Program, Madigan Army Medical Center, Tacoma, WA; 7Department of Internal Medicine, Madigan Army Medical Center, Tacoma, WA
Background
In previous years, efforts to empower T2D self-management via mobile health (mHealth) included portals, text messaging, collection of biometric data, electronic coaching, e-mail, and collection of lifestyle information.
Objective
The goal is to use the U.S. Department of Defense's Mobile Health Care Environment (MHCE) to improve patient activation and self-management of T2D in patient-centered medical home settings.
Methods
A user-centered design and a controlled trial was conducted within the U.S. Military Health System in a multisite study. Phase I evaluated preferences concerning the enhancement of the enabling technology. Phase II was a single-blinded, 12-month feasibility study that randomly assigned 240 patients to either the intervention (n=123, provided with mHealth technology and behavioral messages tailored to Patient Activation Measure [PAM] level at baseline) or the control group (n=117, provided with equipment but not messaging). The primary outcome measure was PAM scores. Secondary outcome measures included Summary of Diabetes Self-Care Activities (SDSCA) scores and cardiometabolic outcomes. Generalized estimating equations were used to estimate changes in outcomes.
Results
A total of 229 patients were included in the final sample. Participants were 61.6% (141/229) male, had a mean age of 62.9 years, mean HbA1c of 7.5%, mean BMI of 32.7, and a mean duration of T2D diagnosis of 9.8 years. At month 12, the control group showed significant improvements compared with the intervention group in PAM scores (control mean 7.49, intervention mean 1.77; P=0.007), HbA1c (control mean −0.53, intervention mean −0.11; P=0.006), and low-density lipoprotein cholesterol (control mean −7.14, intervention mean 4.38; P=0.01). Both groups showed significant improvement in SDSCA, BMI, waist size, and diastolic blood pressure; between-group differences were not statistically significant. Except for patients with the highest level of activation (PAM level 4), patients of the intervention group exhibited significant improvements in PAM scores. For patients with the lowest level of activation (PAM level 1), the intervention group showed significantly greater improvement compared with the control group in HbA1c (control mean −0.09, intervention mean −0.52; P=0.04), BMI (control mean 0.58, intervention mean −1.22; P=0.01), and high-density lipoprotein cholesterol levels (control mean −4.86, intervention mean 3.56; P<0.001). Significant improvements were seen in AM scores, SDSCA, and waist size for both groups and in diastolic and systolic blood pressure for the control group; the between-group differences were not statistically significant. The percentage of participants who were engaged with MHCE for ≥50% of days was 60.7% (68/112; months 0–3), 57.4% (62/108; months 3–6), 49.5% (51/103; months 6–9), and 43% (42/98; months 9–12).
Conclusions
Improvement in PAM scores and outcomes in both the intervention group and control group indicated mixed results. Structural design issues may have hampered the influence of tailored behavioral messaging within the intervention group.
Comment
This study from the U.S. Department of Defense (DOD) is in the exemplary tradition of other DOD and U.S. Veterans Administration studies addressing improving patient engagement in self-management. The authors present not only their intervention and its outcomes, but also discuss how the intervention was created and how they included input from prospective participants in its development (a process called user-centered design). Increased patient activation, as measured by the self-reported patient activation measure (PAM), was the main outcome of the intervention. Increased activation has been demonstrated to correlate with improved clinical outcomes, increased preventative care, and overall lower healthcare-related cost. The PAM survey is associated with self-management behaviors, medication adherence, patient satisfaction, and quality of life. Participants were randomized to intervention or control. The intervention group (n=123) received mHealth technology and behavioral messages tailored to increase the PAM level from baseline, and the control group (n=117) received equipment but no messaging. The changes in PAM and clinical outcomes were mixed without a consistent change across the PAM or clinical measures between the two groups. While this study didn't demonstrate significant outcomes, it nonetheless has a lot to teach us regarding how to do a high-quality study. It is also not surprising that an intervention that is primarily based on providing extra messages, no matter how well written they are, and in the context of clinical care, would be able to change long-time habits and behaviors required to improve outcomes in patients with type 2 diabetes and other chronic conditions.
Educational intervention using WhatsApp on medication adherence in hypertension and diabetes patients: a randomized clinical trial
Sartori AC1, Rodrigues Lucena TF2, Takáo Lopes C3, Picinin Bernuci M4, Yamaguchi MU4
1Cesumar, University Center of Maringá, Paraná, Brazil; 2Department of Fundamentals of Education, UEM, State University of Maringá, Paraná, Brazil; 3Department of Clinical and Surgical Nursing, Paulista School of Nursing, UNIFESP, Federal University of São Paulo, Brazil; 4Health Promotion Graduate Program, Cesumar, University Center of Maringá and ICETI, Cesumar Institute of Science, Technology, and Innovation, Paraná, Brazil
Background
Patients with diabetes mellitus and hypertension frequently do not adhere to pharmacological therapy. This low adherence to pharmacological therapy is a challenge worldwide.
Introduction
This research studied the effects of using WhatsApp messaging for the purposes of medication adherence in patients with hypertension and diabetes.
Materials and Methods
A total of 403 patients with diabetes and/or hypertension who had enrolled in the 33 Basic Health Units of Maringá-PR, Brazil, participated in a randomized clinical trial. The patients were randomly assigned to either the intervention group (n=203), which received the usual care (multiprofessional educational appointments according to each unit schedule) plus 55 audio, image, or text messages via WhatsApp about healthcare promotion, with an emphasis on medication adherence, or the control group. The control group (n=200) received only the usual care. Medication adherence, as measured by the Morisky-Green Test, was compared through the chi-square test after 16 weeks. Relative risk (RR) was used as a measure of effect size.
Results
After the follow-up period of 4 months, 67.5% of the patients in the intervention group were adherent versus 58.5% in the control group (RR: 1.15, 95% confidence interval=0.99–1.34, P=0.077).
Discussion
Although the effect of the intervention was not statistically significant, there was a clinically significant impact associated with a 15% increase in medication adherence.
Conclusion
Given the complexity of adherence to the use of antidiabetic and antihypertensive medications, educational interventions using WhatsApp could be useful as a reinforcement to increase adherence to medication.
Comment
This work is an example of the growing number of interventions that are leveraging health-related text messaging. In the present study, 403 individuals with diabetes or hypertension were randomized to receive either the usual care or the intervention, which consisted of the usual care plus 55 audio, image, or text messages via WhatsApp over a 4-month period to promote self-management behaviors, with an emphasis on medication adherence. This study showed a trend toward improved medication adherence (15%) but was not statistically significant. Most notable about this study is that while text messaging alone is not the “secret sauce” to improving engagement with self-management, it does tend to have a positive impact. Future studies should consider adding text messaging as a “boost” to other effective interventional approaches. We must also remember that not all text messaging software platforms are created equal. Any participants who were not already using WhatsApp had to change their behavior regarding mobile app usage on their phone. Whether such an intervention could be more effective if it used the text messaging app already used by the individual on a daily basis remains to be determined. How many mobile apps do you have on your phone? And how many do you actually use on a daily basis?
A highly tailored text and voice messaging intervention to improve medication adherence in patients with either or both hypertension and type 2 diabetes in a UK primary care setting: feasibility randomized controlled trial of clinical effectiveness
Kassavou A1, Mirzaei V1, Brimicombe J1, Edwards S1, Massou E1, Prevost AT2, Griffin S1, Sutton S1
1The University of Cambridge, Cambridge, UK; 2Imperial College London, Cambridge, UK
Background
A highly tailored digital intervention to support medication adherence has not yet been assessed, nor has the possibility of supporting clinical effectiveness as an adjunct to the primary care setting been evaluated.
Objective
To determine the behavioral efficacy of a highly tailored digital intervention to support medication adherence, as well as the feasibility of its clinical effectiveness, in patients with either or both hypertension and type 2 diabetes, were the aims of this trial. It also analyzed quality of life and mechanisms of behavior change, as well as intervention fidelity, engagement, and satisfaction.
Methods
Two parallel groups participated in a multicenter, individually randomized controlled trial. They consisted of an intervention group that received a highly tailored text message and interactive voice response for a total of 12 weeks and a control group that received the usual care. Adherence to medication was quantified using self-reports and assessor-blinded practice records of a repeat prescription. Nurses blinded to group allocation assessed systolic blood pressure and glucose levels during practice visits at 3-month follow-ups. Questionnaires obtained data to assess intervention mechanisms of action and satisfaction, and digital log files captured data to evaluate fidelity and engagement.
Results
A total of 135 nonadherent patients (62/135, 46% female; 122/135, 90.3%; aged over 50 years) were randomly allocated to the intervention (n=79) or the control group (n=56); 13% (18/135) of patients were lost at follow-up. Adherence to medication was significantly improved in the intervention group compared with the control group (t116=2.27; P=0.02, two-tailed). Systolic blood pressure was 0.6 mmHg (95% CI −7.423 to 6.301) and hemoglobin A1c was 4.5 mmol/mol (95% CI −13.099 to 4.710) lower in the intervention group compared with the control group. Changes in intentional nonadherence and nonintentional nonadherence explained the improvements in medication adherence in the intervention group (beta=0.074, SE=0.464; P=0.04) but not in the control group (beta=0.00, SE 1.35; P=0.37). The intervention had 100% fidelity, a median of 12 days of engagement, and 76% overall satisfaction.
Conclusions
This trial is the first of its kind to take place in the United Kingdom. It showed that, among nonadherent patients with either or both hypertension and type 2 diabetes, a highly tailored digital intervention was effective at improving treatment adherence and feasible to obtain clinically meaningful outcomes. Changes in intentional and nonintentional nonadherence predicted the improvements in adherence to medication. The intervention showed high fidelity, engagement, and satisfaction. Ongoing research using a rigorous design is necessary to determine the clinical effectiveness and cost-effectiveness of the intervention in primary care.
Comment
Persons with chronic illnesses like diabetes or hypertension have a complex relationship with their disease, and many factors influence medication adherence. The present study evaluated the impact of a highly tailored digital intervention on medication adherence in 135 individuals who had either or both type 2 diabetes and hypertension and who exhibited suboptimal adherence at baseline. The intervention included highly tailored text messages and an interactive voice response system. A 12-week randomized controlled trial revealed significant improvements in medication adherence, but no improvements in systolic blood pressure or hemoglobin A1c. The intervention was generally acceptable to participants, but participants receiving the intervention displayed only a modest improvement in medication adherence. The article does a nice job of highlighting intentional vs nonintentional nonadherence, but it highlights something even more important. It is challenging to modify human behaviors, and more intensive communications strategies can only be so effective. Perhaps we need intensive communications strategies to be delivered as just-in-time adaptive interventions (JITAIs), with AI and real-time data to drive the moments and frequency of communication. Or perhaps we need to incentivize individuals who exhibit low medication adherence via behavioral economic interventions. How those interventions are reimbursed remains an open question, so there is much work to be done.
Understanding the adoption and diffusion of a telemonitoring solution in gestational diabetes mellitus: qualitative study
Khalil C
Paris Descartes University, Paris, France
Background
To improve pregnancy outcomes, women with gestational diabetes mellitus (GDM) require regular follow-ups and overall management to normalize maternal blood glucose. Advancements in the digital field have allowed telemedicine to gain popularity over traditional healthcare in various medical fields. Telemonitoring solutions seem to improve women's quality of life and enhance self-management of GDM.
Objective
This study aims to understand, from the perspective of patients and healthcare professionals (HCPs), what drives the adoption and diffusion of a telemonitoring solution (myDiabby) when telemonitoring is not compensated like traditional follow-ups.
Methods
The study took place at 12 diabetes services in France that used myDiabby for monitoring and managing patients with GDM. For collecting and analyzing data, a qualitative research approach was implemented. A total of 20 semistructured interviews took place with HCPs from different health structures in France, and 15 semistructured interviews were conducted with patients who had been using myDiabby. Data were evaluated using a thematic analysis approach.
Results
Different determinants should be considered when adopting an innovative health technology. By drawing on the diffusion of innovation theory, a set of factors associated with the technology (the relative advantages, compatibility, ease of use, testability, and observability of the telemedicine platform) has been identified as affecting the adoption and diffusion of telemonitoring solutions in French diabetes services. In addition, data analysis shows a set of environmental factors (the demographic situation of HCPs, the healthcare access in rural communities, and the economic and political context in France) that also influences the spread and adoption of telemonitoring systems in French hospitals.
Conclusions
Even though telemonitoring activities are still not compensated as traditional follow-ups, many French HCPs support and encourage the adoption of telemonitoring systems in GDM. As for patients, telemonitoring systems are perceived as a useful and easy way to monitor their GDM. This study contributes to recognizing the value of telemonitoring interventions in managing GDM and considering the expansion of telemonitoring to other chronic conditions.
Comment
The present study represents a qualitative evaluation of patients' and healthcare providers' perspectives on the drivers of diffusion and adoption of the MyDiabby telemonitoring solution. The article implements and provides readers an easy-to-comprehend framework (diffusion of innovation theory) for understanding the technology-intrinsic and environmental factors that influence adoption of a telemonitoring solution among French hospitals. What is notable about this article is that, while it focuses on the care of gestational diabetes, it provides a roadmap for thinking about the dissemination of telemonitoring software for diabetes in general. Indeed, MyDiabby has begun to expand its focus in France to include type 1 and type 2 diabetes. As various diabetes data-sharing and telemonitoring software platforms compete for attention in clinics globally, these factors should be considered by each developer in order to achieve success. Developers of telemonitoring solutions should also not presume that drivers of adoption are similar in every country or care system.
Demonstrated cost-effectiveness of a telehomecare program for gestational diabetes mellitus management
Lemelin A1, Paré G2, Bernard S1, Godbout A1
1Endocrinology Division, Medicine Department, Centre de Recherche du Centre hospitalier de l'Université de Montréal (CRCHUM), Montreal, Canada; 2Research Chair in Digital Health, HEC Montreal, Montreal, Canada
Background
Prevalence of GDM has increased steadily in recent years. Pregnant women with GDM are at risk for obstetrical and neonatal complications and require close multidisciplinary follow-up, which implies a significant use of hospital resources.
Methods
A prospective noninferiority and controlled clinical trial was designed. The telehomecare (THCa) initiative is a clinical remote patient management project for women with GDM. The main objective was to evaluate the cost-effectiveness of THCa by assessing the direct costs, including the related reduction in medical visits. Secondary outcomes were to evaluate the impact of THCa on diabetes control, GDM-related complications, and patient satisfaction.
Results
A total of 161 women were assigned to either an intervention group provided with a THCa system for transmission and online analysis of capillary glucose data (n=80) or a control group receiving usual care in the clinic (n=81). A decrease in medical visits by 56% (P < 0.001) in the THCa group was observed. There was no difference between the two groups in diabetes control or maternal and fetal complications. However, results showed a 10-fold increase in nursing interventions in the THCa group (mainly by phone calls and e-mails). Satisfaction with care was high. Direct cost analysis revealed savings of 16% in patients followed by THCa compared with the control group.
Conclusion
THCa monitoring significantly decreases medical visits and direct costs in GDM women without compromising pregnancy outcomes, quality of care, or patient satisfaction. THCa was shown to be cost-effective despite placing an additional burden on nursing time.
Comment
Whether remote patient monitoring and telehealth, when used together, are noninferior to in-person care is an important question in healthcare generally. Another important question is whether remote patient monitoring and telehealth are, in fact, cost-superior to in-person care. These questions are especially important in the care of individuals with gestational diabetes, in whom the potential for maternal or fetal health complications is significant. These authors evaluated the cost effectiveness of a remote telehomecare intervention (THCa) in 161 women. They found no difference in diabetes control or fetal complications, yet the authors identified a 16% cost savings among the group who received THCa. It appears that the cost savings can be accounted for by the “conversion” of in-person care episodes to an increased frequency of nurse contacts via phone or e-mail. Clinicians and healthcare administrators find it difficult to change their procedures. But as the world grapples with a pandemic that has forced rapid migration to direct-to-consumer telehealth approaches, we have a unique opportunity to increase our understanding of the efficacy and cost effectiveness of telehealth-driven approaches to care. Let's hope we see more studies like this in the coming years.
An android smartphone-based randomized intervention improves the quality of life in patients with type 2 diabetes in Mysore, Karnataka, India
Sunil Kumar D1, Prakash B1, Subhash Chandra BJ2, Kadkol PS3, Arun V3, Thomas JJ1
1Department of Community Medicine, JSS Medical College, JSS Academy of Higher Education and Research, Mysuru, India; 2Department of General Medicine, JSS Medical College, JSS Academy of Higher Education and Research, Mysuru, India; 3Department of Information Science & Engineering, JSS S&T University (Formally SJCE), Mysuru, India
Background
Diabetes mellitus is a significant public health burden, as it is associated with severe complications and morbidity. The need for regular monitoring and adherence to treatment and lifestyle changes have a high impact on the quality of life of the patients. Through a randomized field trial, this study aims to evaluate the effects of smartphone-based lifestyle modification intervention in the quality of life of patients with type 2 diabetes.
Methods
Patients attending the outpatient department of a tertiary care hospital in Mysuru City participated in a randomized field trial. For a period of 6 months, from April 2019 to September 2019, a mobile application named DIAGURU was used, mainly focusing on lifestyle modification and medication management. A total of 150 patients were part of the intervention group while another 150 participants served as controls. Quality of life was assessed using the WHO QOL BREF questionnaire at the beginning of the study and after 6 months.
Results
After 6 months, positive changes in quality of life were significantly higher in the intervention group compared with the nonintervention group. After the intervention, the differences in the change in scores of quality of life for participants in intervention and nonintervention groups were statistically significant in all four domains, with a P value <0.001.
Conclusion
The evidence from this study suggests that such technological approaches can be used as a public health measure to improve the quality of life for patients with type 2 diabetes mellitus.
Comment
This study randomized 300 people with type 2 diabetes to either an Android smartphone-based intervention (N=150) or a control (N=150). The main support received by the intervention group was related to diet and medication, including some information and answers to frequently asked questions. Despite the relatively modest amount of support the participants received, the researchers were able to demonstrate improved quality of life in the intervention group. This exciting and promising outcome is very encouraging. A larger, different study is needed to demonstrate improved clinical outcomes in addition to quality of life.
Effectiveness of a social media-based, health literacy–sensitive diabetes self-management intervention: a randomized controlled trial
Kim SH1, Utz S2
1College of Nursing, Research Institute of Nursing Science, Kyungpook National University, Daegu, South Korea; 2Social Media Leibniz-Institut für Wissensmedien, University of Tübingen, Tübingen, Germany
Objective
This study aimed to assess the effects of a social media–based, health literacy–sensitive diabetes management intervention on patient activation, self-care behaviors, and glucose control compared to telephone-based, health literacy–sensitive diabetes management intervention and usual care. Another goal was to determine how patient health literacy influences the effectiveness of health literacy–sensitive diabetes management interventions.
Methods
A total of 151 patients with T2D were randomly assigned to the social media–based or telephone-based, health literacy–sensitive diabetes management interventions or the usual care control. The health literacy–sensitive diabetes management intervention included an initial face-to-face diabetes nurse education with easy-to-read instructive materials, as well as the teach-back method and eight weekly action-planning sessions directed by social media or phone calls for each group.
Results
At the 9-week follow-up, patients with high health literacy showed higher levels of patient activation than those with low health literacy in the control group, but the effect of health literacy was no longer significant when patients were provided with social media–based or telephone–based interventions. Patients who received the telephone-based, health literacy–sensitive diabetes management intervention had a significantly higher score for self-care behaviors than the usual care control group at the 9-weeks follow-up. No other effects for self-care behaviors or glycated hemoglobin were significant at follow-up.
Conclusions
The social media–based, health literacy–sensitive diabetes management intervention was effective at lessening the disadvantages faced by people with low health literacy when attempting to improve self-care activation.
Comment
This is a very important and ground-breaking study, not only because of its solid research design but also because it tested the use of social media services on adults with low health-literacy skills. Note that social media services are defined in the article as “web-based services that allow individuals, communities, and organizations to collaborate, connect, interact, and build a community by enabling them to create, co-create, modify, share, and engage with user-generated content that is easily accessible.” The results demonstrated the social media–based self-management interventions accommodating low health literacy can potentially help people overcome disadvantages associated with low health literacy. Additional studies are needed to demonstrate long-term outcomes that are clinically and economically relevant for these approaches to go to scale. Hopefully, this can happen soon so we can make a dent in poor outcomes, especially for people with diabetes from underserved communities.
Supporting good intentions with good evidence: how to increase the benefits of diabetes social media
Reidy C1,2, Klonoff DC3, Barnard-Kelly KD2,4
1Wessex CLAHRC, School of Health Sciences, Faculty of Environmental and Life Sciences, University of Southampton, Southampton, UK; 2BHR Limited, Portsmouth, UK; 3Mills-Peninsula Medical Center, San Mateo, CA; 4Bournemouth University, Bournemouth, UK
Abstract
For millions of people with diabetes, social media offers a platform for easily accessible, relevant health information as well as emotional and practical support at the touch of a button. Therein, however, lies a challenge. The accuracy and reliability of the information is often unknown and unverified, not all interactions are deemed supportive—practically or emotionally—and not all members of society have equitable access.
Reliance on web-based health information has generated concerns about patients' ability to accurately evaluate the credibility of online sources as well as the possible detrimental effects on personal well-being and patient–provider relations. Furthermore, there are rising digital disparities for certain subpopulations. Such concerns apply to where and how healthcare professionals should refer patients to or engage with patients in terms of platforms of online support.
Both within and outside of the health arena there is little doubt about the popularity of social media, but there are also concerns. This article reviews five key areas correlated with social media use in people living with diabetes and presents possible concerns moving forward. The authors focus on (1) social media as a platform for information and support; (2) social media interactions that are not supportive; (3) lessons from the Diabetes Online Community (DOC); (4) concerns about accuracy, reliability, and accessibility of information; and (5) differing priorities of healthcare professionals and patients.
Comment
This article summarizes and addresses several issues relevant to the use of social media to improve outcomes for people with diabetes. The authors highlight key issues related to the relevance, accuracy, and acceptability of peer-to-peer support provided through social media. They highlight potential positives and negatives, giving the reader a thoughtful framework to help determine the key elements for successful social media approaches. There is no doubt that social media is here to stay, Clinicians need to get used to it, and more importantly, be prepared to contribute to it and refer patients to vetted sites so that social media's impact can become even greater.
Can parental engagement in social media enhance outcomes of an online healthy lifestyle program for preschool-aged children?
Hammersley ML1, Okely AD1, Batterham MJ2, Jones RA1
1Early Start, Faculty of Social Sciences, University of Wollongong, Wollongong, Australia; 2Statistical Consulting Service, National Institute for Applied Statistics Research Australia, School of Mathematics and Applied Statistics, University of Wollongong, Wollongong, Australia
Aim
The aim of this study was to assess parental engagement, child BMI, and secondary outcomes from the social media component of an online healthy lifestyle program for parents of preschool-aged children.
Methods
Participants of the intervention group were given access to an online program and Facebook group. Data were collected at baseline and follow-up data collected at 3 and 6 months. Comments and posts on Facebook were used to establish total active engagement.
Results
A high level of Facebook group membership was observed and most parents actively engaged at least once. Between modules and cohorts, there were varying levels of engagement, but overall, engagement was modest. User acceptability of the Facebook group was lower than anticipated. Children of parents in the intervention group who engaged more in the Facebook group (by posting and commenting) exhibited greater sleep duration over time (estimate 1.79, 95% CI 0.42 to 3.17, P=0.01). Children of parents who engaged more in the Facebook group also participated in less moderate- to vigorous-intensity physical activity (estimate −0.14, 95% CI −0.26 to −0.01, P=0.03).
Conclusions
This is one of the first studies on parent-focused healthy lifestyle interventions to include a social media component. Additional research with larger sample sizes and longer duration to further explore the potential of social media in childhood obesity interventions is recommended.
Comment
This substudy was part of a larger study called Time2bHealthy on the impact of adding a private Facebook group to participants in an 11-week, healthy lifestyle intervention for parents of preschool-aged children at risk for overweight/obesity. The children of parents assigned to the Facebook group showed no difference in their weight or their diet but reported more sleep—an outcome that was likely welcomed by the parents. Of course, programs designed to impact preschoolers need to be directed to the parent as well as the child, with age-appropriate lessons and activities. Not an easy task.
Effectiveness of social media (Facebook), targeted mailing, and in-person solicitation for the recruitment of young adult in a diabetes self-management clinical trial
Salvy SJ1, Carandang K2, Vigen CL3, Concha-Chavez A4, Sequeira PA5, Blanchard J3, Diaz J3, Raymond J6, Pyatak EA3
1Research Center for Health Equity, Cedars-Sinai Medical Center, West Hollywood, CA; 2University of California at San Diego, San Diego, CA; 3Chan Division of Occupational Science and Occupational Therapy, University of Southern California, Los Angeles, CA; 4Northern Arizona University, Phoenix, AZ; 5Los Angeles Department of Health Services, Los Angeles, CA; 6Children's Hospital Los Angeles, Los Angeles, CA
Background
Studies to determine effective recruitment strategies for reaching and engaging diverse young adults in diabetes clinical research are needed. This study's goal was to assess the strengths and weaknesses of three recruitment strategies used in diabetes self-management clinical trials: social media advertising (Facebook), targeted mailing, and in-person solicitation of clinic patients.
Methods
Strategies were assessed according to (1) cost-effectiveness (i.e., cost of recruitment/number of enrolled participants), (2) ability to yield participants who would not otherwise be reached by alternative strategies, and (3) likelihood of participants recruited through each strategy to adhere to study procedures. The researchers also analyzed the appeal (overall and among age and gender subgroups) of certain advertisement features on social media.
Results
Overall, the most cost-effective strategy was in-person recruitment of clinic patients, but differences in demographics as well as clinical and psychosocial characteristics of participants that were recruited via various strategies suggest that the combination of these approaches produced a more diverse sample than any one strategy alone. Once successfully enrolled, the participants exhibited no difference in study completion and intervention adherence among those recruited by the varying strategies.
Conclusions
The utility of a recruitment strategy is ultimately defined by its ability to attract people of the target population who are willing to enroll in and complete the study. Leveraging a variety of recruitment strategies seems to generate a more representative sample of young adults, including those who are typically less engaged in diabetes care.
Comment
This article reports on the effectiveness of social media, in-person solicitation, and targeted mailing to successfully recruit young adults (18–30 years old) into diabetes clinical research studies. While this study isn't representative of recruitment into clinical and nonresearch programs, it does provide information about this important but under-studied area. In-person recruitment was the most cost-effective approach, but that isn't practical for population health approaches that are expected to recruit large numbers of participants. Use of Facebook advertisements required the least staff time, an important consideration given limitations in funding. Also, using all three methods yielded a more diverse and representative sample of the targeted population. The challenge of enlisting the right person in the right program during the right time of the person's health or illness journey is a key factor in the success or failure of any intervention. This is all the more difficult given the “competition” from direct-to-consumer approaches that do not have the regulatory or clinical restrictions placed on clinical services.
Randomized controlled trial of OnTrack, a just-in-time adaptive intervention designed to enhance weight loss
Forman EM1, Goldstein SP2, Crochiere RJ1, Butryn ML1, Juarascio AS1, Zhang F1, Foster GD3
1Center for Weight, Eating, and Lifestyle Sciences (WELL Center), Drexel University, Philadelphia, PA; 2Weight Control & Diabetes Research Center, Warren Alpert Medical School of Brown University, Providence, RI; 3Weight Watchers International, New York, NY
Background
Instances of nonadherence to reduced-calorie dietary prescriptions—that is, dietary lapses—are a major challenge for weight management. Just-in-time adaptive interventions (JITAIs) collect and analyze data in real time to deliver tailored interventions at relevant moments and may be well-suited to encourage weight loss by thwarting dietary lapses.
Aims
OnTrack (OT) is a smartphone application (app) that collects data on lapses and triggers of lapses, uses a continuously improving machine-learning model to predict lapse risk, and provides tailored interventions when risks are elevated. This study assessed the efficacy of OT against an active control in aiding weight loss.
Methods
Participants (N=181) with overweight/obesity (mean body mass index [MBMI]=34.32; 85.1% female; 73.5% White) were randomized to receive either the WW (formerly Weight Watchers) Beyond the Scale (BTS) digital program alone or WW plus OnTrack (WW + OT) for 10 weeks. In an unplanned, natural experiment, the WW program switched midway through the trial from BTS to a more flexible program called Freestyle (FS).
Results
A general linear model revealed a treatment condition × diet plan interaction (F[1, 173]=9.68, P=0.002) such that OT exhibited greater efficacy only among those receiving BTS (weight loss MWW + OT=4.7%, standard error [SE]=0.55 versus MWW=2.6%, SE=0.80). Compared to BTS WW + OT participants reported considerably higher satisfaction with the intervention than with FS. In addition, engagement was higher and algorithm accuracy was superior.
Conclusion
These results offer qualified support for OT overall and for machine learning–powered JITAIs in general that facilitate weight loss by predicting and preventing dietary lapses.
Comment
This randomized controlled experiment compared traditional Weight Watchers support to Weight Watchers with OnTrack—a smartphone application that collects data on lapses and triggers of lapses. The OnTrack intervention is determined by machine learning to provide just-in-time messages that inform and support participants by delivering tailored interventions when risk of a lapse in health-promoting behaviors is elevated—an exciting new approach with applicability across a variety of behaviors, such as unhealthy eating, smoking/drug use, non-adherence with medications, and more. While the clinical outcomes weren't particularly impressive, I expect that we will be seeing more studies like this one. I certainly hope so.
Dietary self-monitoring through calorie tracking but not through a digital photography app is associated with significant weight loss: the 2SMART pilot study—a 6-month randomized trial
Dunn CG1,2, Turner-McGrievy GM3, Wilcox S4, Hutto B4
1Department of Health Policy and Management, Harvard T.H. Chan School of Public Health, Boston, MA; 2Department of Health Promotion, Education, and Behavior, and the Prevention Research Center, Arnold School of Public Health, University of South Carolina, Columbia, SC; 3Department of Health Promotion, Education, and Behavior, Arnold School of Public Health, University of South Carolina, Columbia, SC; 4Prevention Research Center, and Department of Exercise Science, Arnold School of Public Health, University of South Carolina, Columbia, SC
Background
Among behavioral interventions, dietary self-monitoring (DSM) of foods and beverages is correlated with weight loss; however, DSM can be troublesome, and adherence may decrease over time. Novel methods of DSM that include apps to track food using photographs may decrease inconvenience, increase DSM adherence, and improve weight loss.
Objective
The aim was to assess a mobile photo DSM app compared to a calorie-tracking DSM app on tracking frequency and weight loss in a remotely delivered behavioral weight-loss intervention.
Methods
This was a 6-month (October 2016 to April 2017) randomized trial of adult participants (n=41) from South Carolina who were considered overweight or obese (body mass index 25 to 49.9). Participants received remotely delivered twice-weekly behavioral weight-loss podcasts and tracked their diet using a calorie-tracking DSM app (calorie group) or a photo DSM app (photo group). The key outcomes measured were the number of days diet was tracked, number of podcasts downloaded, and amount of weight changed at 6 weeks and 6 months.
Results
No differences were shown between groups for the number of days that diet was recorded (P=0.18), which was low overall (<30% of days) but was statistically significantly and strongly correlated with weight change for all participants pooled (r=0.63; P<0.001) and for the calorie tracking group (r=0.70; P=0.004), but not the photo tracking group (r=0.51; P=0.06). Participants in both groups had significant weight loss at 6 months (photo group, −2.5±0.9 kg; P=0.008; calorie group −2.4±0.9 kg; P=0.007), with no differences between groups at either 6 weeks (P=0.66) or at 6 months (P=0.74).
Conclusions
Frequency of DSM was significantly correlated with overall weight loss for participants using a calorie DSM app but not a photo DSM app. DSM was low regardless of group, and weight loss was significant, although minimal. Improving user engagement with any DSM may be relevant to increasing self-monitoring and improving weight loss.
Comment
As diabetes self-management and physical activity wearable devices increasingly develop the ability to passively stream data to cloud repositories, dietary self-monitoring remains encumbered by the highly manual, burdensome nature of the task. The fact that individuals disengage from manual dietary tracking fairly quickly highlights the need for improved methods that reduce the user burden and improve longitudinal engagement. This study evaluated the impact of remote food photography compared to calorie tracking on user engagement with continued dietary self-monitoring over 6 months. The participant population included 41 adults in South Carolina, who participated in an intervention involving twice-weekly podcasts and dietary self-monitoring; participants were randomized to a remote food photography group versus a calorie tracking app group. The authors found that remote food photography was not associated with increased engagement with dietary tracking, and that participants self-monitored diet on <30% of days. Weight loss correlated with the number of days that diet was recorded in the calorie-tracking group but not the remote food photography group. It seems that taking pictures of one's meals is not quite as compelling as Facebook and Instagram would have us think.
A systematic review to assess the effectiveness of technology-based interventions to address obesity in children
McMullan M1, Millar R1, Woodside JV2
1School of Medicine, Dentistry and Biomedical Sciences, Queen's University Belfast, Belfast, UK; 2Institute for Global Food Security (Centre for Public Health), Belfast, UK
Background
Hypertension, hyperlipidemia, cardiovascular disease, and type 2 diabetes are just some of the comorbidities associated with childhood obesity. Negative prejudice and social marginalization are also correlated with childhood obesity and can additionally affect a young person's social, emotional, and mental health. Given the prevalence of overweight and obese children globally, it is crucial that effective interventions are established. Since children are receptive to information conveyed via digital means, the use of technology may play an essential role in interventions to reduce childhood obesity. This systematic review aimed to assess and critically appraise the literature published to date in relation to the effectiveness of technology-based interventions, implemented as secondary prevention, in addressing childhood obesity.
Methods
Using Medline and Embase, an electronic search strategy was undertaken that included publications up to July 12, 2018. Randomized controlled trials, published only in the English language, evaluating the effectiveness of technology-based interventions on weight-related outcomes in children, aged 8 to 18, were included.
Results
A total of 11 studies met the inclusion criteria out of an initial search of 1012 studies. They were evaluated for methodological quality using the Cochrane Risk of Bias Tool for Randomized Controlled Trials and were analyzed using a narrative approach. The findings showed a limited potential of technology-based interventions, employed as secondary prevention, to address childhood obesity. Of the 11 studies reviewed, 3 (27%) showed a positive relationship between technology-based interventions and weight-related outcomes in overweight or obese children.
Conclusions
This review suggests that technology-based interventions, which primarily include active video games as well as Internet or web-based interventions and mobile phone communications, may, with additional research, have the potential to impact positively on weight-related outcomes. It is difficult to determine the degree of efficacy of these technology-based interventions, as only two databases of only English-language articles were searched. Additionally, the studies chosen showed a lack of high-quality evidence. The lack and heterogeneity of studies with technology-based interventions is another limitation.
Comment
This article represents a new systematic review of technology-based interventions for secondary prevention of pediatric obesity. The article has the added benefit of summarizing the findings of several prior systematic reviews addressing primary prevention of pediatric obesity. Eleven of 1012 articles met the authors' inclusion criteria, and 3 of the studies showed a positive relationship between technology-based interventions and weight-related outcomes in children with obesity. The authors evaluated interventions involving active video games, Internet-based treatment, and mobile phone communications; 2 of 5 studies involving active video games showed a benefit, and 1 out of 5 articles evaluating Internet-based treatment showed a benefit. The lone study that was evaluated involving phone communications, text messages, or e-mails to augment a community-based weight program failed to show a benefit. My conclusion? We had better get off the couch and start iterating these potentially effective interventional strategies. Children coping with obesity really need researchers to be active (in this area of investigation).
Type 1 diabetes self-management with game-based interventions for pediatric and adolescent patients
Rewolinski JA1, Kelemen A1, Liang Y2
1Department of Organizational Systems and Adult Health, University of Maryland, Baltimore, MA, 2Department of Family and Community Health, University of Maryland, Baltimore, MA
Background
T1D has a peak diagnosis between the ages of 10 and 14 and requires intense lifestyle changes. Self-management is crucial for sufficient metabolic control to prevent acute and long-term complications. Yet common approaches to diabetes self-management education, such as lectures or pamphlets, lead to lack of sufficient knowledge, engagement, and clinical outcomes. On the other hand, game-based learning has led to increased motivation, engagement, and productivity overall, with notable increases in self-management of chronic diseases in children.
Aims
This article aims to review literature concerning the impact on self-management knowledge, behavior, and engagement of the interventions of serious games and gamification for children and adolescents with type 1 diabetes.
Results
Nine studies were reviewed and showed statistically significant differences in knowledge, behavior, and engagement in response to the game-based interventions. Knowledge outcomes were most significant in serious game interventions, while behavioral outcomes were predominantly found in gamification/serious game combination interventions. Findings also revealed inconsistent use of theories for game development and moderate- to low-quality evidence across studies.
Conclusions
While the nine studies reviewed strongly indicate the potential of game-based tools to significantly improve type 1 diabetes self-management care, additional studies with expanded and more rigorous parameters are recommended before an outright change in practice should be considered.
Comment
This article reviews nine studies, evaluating the impact of serious games and gamification (game-based incentives) on supporting self-management knowledge, behavior, and engagement among youth and adolescents with type 1 diabetes. The review concludes that knowledge, behavior, and engagement do increase in response to serious games and gamification. The authors criticize past studies for lack of consistency in applying theories of game development and for limitations in study design. What should be obvious from this review is that the field of game design as it applies to serious games and gamification in healthcare remains disconnected from the fields of behavioral psychology, diabetology, and clinical trial design. Digital health interventions should be evaluated with the same rigor as drugs and devices, yet common clinical trial designs fail to adequately address the multicomponent nature of many digital health interventions, including serious games and gamification strategies in healthcare. The field needs to demand the implementation of more rigorous and innovative clinical trial designs evaluating serious games and gamification for diabetes self-management. I say we “gamify” clinical trial design by having researchers play a game in which the most grant dollars go to the group(s) that create the most engaging and effective games for healthcare.
Effects of active video games on children's psychosocial beliefs and school day energy expenditure
Gao Z1,2, Pope ZC3, Lee JE4, Quan M1
1Department of Sport Rehabilitation, School of Kinesiology, Shanghai University of Sport, Shanghai, China; 2School of Kinesiology, University of Minnesota-Twin Cities, Minneapolis, MN; 3School of Public Health, University of Minnesota-Twin Cities, Minneapolis, MN; 4Department of Applied Human Sciences, University of Minnesota, Duluth, MN
Aim
The goal of this study was to assess the effects of active video games (AVGs) on children's school-day energy expenditure (EE) and physical activity (PA)–related self-efficacy, social support, and outcome expectancy.
Methods
Over the span of 9 months in 2014–2015, 81 fourth-grade students (Xage=9.23 years, SD=0.62; 39 girls) from two urban Minnesota elementary schools participated in this study. At the intervention school, a once-weekly, 50-min AVG intervention was implemented, while the control school continued regular recess. Children's school-day EE (daily caloric expenditure) and mean daily metabolic equivalent (MET) values were estimated via accelerometry, whereas self-efficacy, social support, and outcome expectancy were evaluated with psychometrically validated questionnaires. All measures were completed initially at baseline and then twice more at 4 and 9 months.
Results
The research team noted significant interaction effects for daily caloric expenditure: F(1, 58)=15.8, P<0.01; mean daily MET values, F(1, 58)=11.3, P<0.01; and outcome expectancy, F(1, 58)=4.5, P<0.05. Students involved in the intervention group experienced greater increases in daily caloric expenditure (91 kilocalorie/day postintervention group difference), with the control group showing decreasing daily caloric expenditure over time. The authors noted identical trends for mean daily MET values (0.35 METs/day postintervention group difference). At postintervention, increased outcome expectancy in the control children, but decreased expectancy among intervention children was also observed (1.35 group difference). Finally, authors observed a marginally significant interaction effect for social support, F(1, 58)=3.104, P=0.08, with an increase and decrease in the intervention and control children, respectively. They observed no interaction or main effects for self-efficacy.
Conclusion
Results suggested an AVG intervention added to longitudinal increases in school-day EE and social support compared to the control condition. Research in the future should evaluate how self-efficacy and outcome expectancy might be promoted during school-based AVG interventions.
Comment
The authors perform a really interesting experiment. Students in a fourth-grade classroom at two different elementary schools were assigned, as a group, to either an active video game intervention (via Xbox or Wii) or regular recess during a 9-month school year. The students in the intervention group displayed increases in energy expenditure (91 kcal/day difference) and physical activity intensity (measured via metabolic equivalent of task [MET]; 0.35 METs/day difference); in fact, the control students experienced decreasing energy expenditure over time. Future studies should examine the interaction of self-efficacy, child-centric variables (e.g., executive functioning, peer interactions), social support, and other family or neighborhood factors. Despite the need for additional investigations, these findings provide reason for optimism considering the intervention appeared to have a sustainable effect over 9 months. It would be great if school officials could test-deploy this intervention across school systems, and greater still if they could extend this intervention into the home for all students attending virtual school.
CLOSING REMARKS
Digital therapeutics—what's next?
The future for digital interventions is bright, especially in a COVID-19 and post COVID-19 world.
There has been a near universal acceptance of telemedicine (provider–patient visits) for the management of acute and, in many cases, chronic conditions.
There has also been a major transition from in-person delivered, community-based group education and interventions that are now being delivered digitally (either by distance learning [synchronized, group-based delivery of previously in-person programs] and/or interventions specifically designed to be delivered digitally).
An increasing number of digital therapeutics have demonstrated the required clinical evidence for regulatory approval and payer coverage.
The field is still relatively young but now includes more than startups—some companies have valuations of greater than $1 billion.
The next phase will be about how digital therapeutics can further weave into general care.
Requirements for the digital therapeutic enterprise to be successful include: Improved interventions designed with authentic input from the target population. The development of innovative clinical trial designs that are better suited to evaluating the various components of digital health interventions, which are often multicomponent in nature. Development of research methodologies and studies that demonstrate outcomes relevant to people with diabetes, providers, payers, community-based organizations, and governments. Creation of more and improved collaborations between developers of digital therapeutics and plans, providers, pharma/device companies, and employers. Enhanced focus not only on the outcomes for those participants who enroll in a program but also on how to cost-effectively identify and recruit the right person for the right intervention at the right time in their health journey. Creation of a digital therapeutic ecosystem capable of not only offering high-quality interventions but also able to be integrated into the core workflows used to provide direct medical health services. Development of new/improved business models that not only encourage innovation but are also able to sustainably bring the interventions to scale.
Footnotes
Author Disclosure Statement
NK works for and has equity interest in Canary Health, LLC. NK's son is an employee of and has equity interest in WW, the subject of the article regarding Ontrack. MAC is employed by Glooko, Inc. He has received consulting fees from Eli Lilly and Medtronic. He has received research support from Dexcom and Abbott Diabetes Care. EM has no competing financial interests.