Abstract
Introduction
Diabetes digital health includes electronic health (eHealth, which can be web-based or computer software–based), mobile health (mHealth, which also includes digital therapeutics and digital companion apps), as well as telehealth, including remote monitoring of patients. This year's PubMed search for and review of articles addressing diabetes digital health yielded 22 publications that should be of significant interest to the field. The publications this year can best be summarized by the numbers: Five meta-analyses or systematic reviews Two implementation science approaches Six health economic (mostly cost-effectiveness) analyses Eight randomized controlled trials (either initial reports or secondary analyses) Five pediatric studies of type 1 diabetes (T1D), type 2 diabetes (T2D), prediabetes, or obesity One study on older adults with T2D One study addressing rural compared with urban populations Four studies of digital solutions to deliver the Diabetes Prevention Program Six platforms that can best be described as participatory health technologies or collaborative care platforms Two studies specifically focused on digital therapeutics/digital companions, though many of the mobile phone–based digital health solutions described herein can be considered in those categories Three studies addressing aspects of remote patient monitoring One study of a web-based diabetes education platform, with massively open online courses (MOOC)
The range of topics and the increasing number of higher quality articles indicate an actively maturing field. A title- and abstract-based PubMed search for “diabetes” and “digital health” revealed 274 articles in 20 months from January 2022 in contrast with 193 in all of 2020 and 2021. A similar search focused on “mHealth” revealed 225 articles in the 20 months from January 2022 versus 235 in all of 2020 and 2021. When one searches for “telehealth”-related articles using the same strategy, one finds 335 articles published to date in 2022 and 2023 versus 282 in all of 2020 and 2021. Finally, a search for “remote patient monitoring” and “diabetes” reveals 40 published articles so far in 2022 and 2023 versus 24 in 2020 and 2021. This article focuses on the 23 articles of greatest interest published from July 1, 2022, to June 30, 2023, across this highly varied discipline.
A few concepts of interest emerge in this year's selected articles, including the value of comparing the effectiveness of different forms (aka vehicles for delivery) of digital intervention and the concept of the digital phenotype. Increasing attention is being paid to the important role that bona fide “digital therapeutics” and “digital companions” can play in improving disease outcomes. A digital therapeutic is software as medical device that is supported by evidence to prevent, manage, or treat a medical disorder whereas a digital companion works alongside a medication or treatment to increase patient engagement in the treatment. An example of a digital therapeutic is a cognitive behavioral therapy–based mobile intervention for T2D. Note that digital therapeutics can include some digital companion-like functions (promoting engagement with taking medications). By contrast, a mobile insulin titration application, which is only designed to aid in dose optimization of either long- or short-acting insulins, serves as an example of a digital companion. Health economic analyses continue to reveal the noninferiority or even superiority of digital health solutions relative to nondigital care modalities. This article presents the 22 articles of greatest interest across these highly varied disciplines.
Key Articles Reviewed
Siopis G, Moschonis G, Eweka E, Jung J, Kwasnicka D, Asare BY, Kodithuwakku V, Willems R, Verhaeghe N, Annemans L, Vedanthan R, Oldenburg B, Manios Y, for theDigiCare4You Consortium
Smith DH, O'Keeffe-Rosetti M, Fitzpatrick SL, Mayhew M, Firemark AJ, Gruß I, Nyongesa DB, Smith N, Dickerson JF, Stevens VJ, Vollmer WM, Fortmann SP
De-Jongh González O, Tugault-Lafleur CN, Buckler EJ, Hamilton J, Ho J, Buchholz A, Morrison KM, Ball GD, Mâsse LC
Zhang X, Zhang L, Lin Y, Liu Y, Yang X, Cao W, Ji Y, Chang C
Kerr D, Edelman S, Vespasiani G, Khunti K
Sapanel Y, Tadeo X, Brenna CTA, Remus A, Koerber F, Cloutier LM, Tremblay G, Blasiak A, Hardesty CL, Yoong J, Ho D
Lee EY, Cha SA, Yun JS, Lim SY, Lee JH, Ahn YB, Yoon KH, Hyun MK, Ko SH
Hsia J, Guthrie NL, Lupinacci P, Gubbi A, Denham D, Berman MA, Bonaca MP
Tornvall I, Kenny D, Wubishet BL, Russell A, Menon A, Comans T
Alexandrou C, Henriksson H, Henström M, Henriksson P, Delisle Nyström C, Bendtsen M, Löf M
Chang A, Gao MZ, Ferstad JO, Dupenloup P, Zaharieva DP, Maahs DM, Prahalad P, Johari R, Scheinker D
Crossen SS, Romero CC, Lewis C, Glaser NS
Dupenloup P, Pei RL, Chang A, Gao MZ, Prahalad P, Johari R, Schulman K, Addala A, Zaharieva DP, Maahs DM, Scheinker D, on behalf of the 4T Research Team
Cunningham SG, Stoddart A, Wild SH, Conway NJ, Gray AM, Wake DJ
Heald AH, Roberts S, Albeda Gimeno L, Gilingham E, James M, White A, Saboo A, Beresford L, Crofts A, Abraham J
Fitzpatrick SL, Mayhew M, Rawlings AM, Smith N, Nyongesa DB, Vollmer WM, Stevens VJ, Grall SK, Fortmann SP
Stevens S, Gallagher S, Andrews T, Ashall-Payne L, Humphreys L, Leigh S
Graham SA, Auster-Gussman LA, Lockwood KG, Branch OH
Egilsson E, Bjarnason R, Njardvik U
Wannheden C, Åberg-Wennerholm M, Dahlberg M, Revenäs Å, Tolf S, Eftimovska E, Brommels M
Lee JJN, Abdul Aziz A, Chan ST, Raja Abdul Sahrizan RSFB, Ooi AYY, Teh YT, Iqbal U, Ismail NA, Yang A, Yang J, Teh DBL, Lim LL
Michaud TL, Wilson KE, Katula JA, You W, Estabrooks PA
Effectiveness, Reach, Uptake, and Feasibility of Digital Health Interventions for Adults with Type 2 Diabetes: A Systematic Review and Meta-analysis of Randomised Controlled Trials
Siopis G1,3, Moschonis G1, Eweka E4, Jung J5, Kwasnicka D6, Asare BY7, Kodithuwakku V6, Willems R8, Verhaeghe N8,9, Annemans L8, Vedanthan R4, Oldenburg B2,6, Manios Y10,11, for theDigiCare4You Consortium
1Department of Food, Nutrition and Dietetics, School of Allied Health, Human Services and Sport, La Trobe University, Melbourne, VIC, Australia; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 2Academic and Research Collaborative in Health, La Trobe University, Melbourne, VIC, Australia; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 3Institute for Physical Activity and Nutrition, Deakin University, Geelong, VIC, Australia; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 4Department of Population Health, NYU Grossman School of Medicine, New York, NY; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 5Maternal, Child and Adolescent Health Program, Burnet Institute, Melbourne, VIC, Australia; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 6NHMRC CRE in Digital Technology to Transform Chronic Disease Outcomes, Baker Heart and Diabetes Institute, Melbourne, VIC, Australia; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 7Curtin School of Population Health, Curtin University, Perth, WA, Australia; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 8Department of Public Health and Primary Care, Faculty of Medicine, Ghent University, Ghent, Belgium; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 9Research Institute for Work and Society, HIVA KU Leuven, Leuven, Belgium; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 10Department of Nutrition and Dietetics, School of Health Science and Education, Harokopio University, Athens, Greece; Hellenic Mediterranean University Research Centre, Heraklion, Greece; 11Institute of Agri-food and Life Sciences, Hellenic Mediterranean University Research Centre, Heraklion, Greece
Nine in 10 people worldwide own a phone with features and can receive SMS text messages, and four in five people have access to a smartphone. Smartphone applications are already available for diabetes management. Because comparisons have been sparse of the effectiveness and implementation of the different modes of digital health interventions in type 2 diabetes (T2D), this study compared the effectiveness of SMS texting, smartphone applications, and website-based interventions for improving glycemia in adults with T2D and reports on their reach, uptake, and feasibility.
Methods
This systematic review and meta-analysis examined the effectiveness of digital health interventions in reducing glycated hemoglobin A1c (HbA1c) in adults with T2D. Randomized controlled trials (RCTs) published in English were located via searches of CINAHL, Cochrane Central, Embase, MEDLINE, and PsycInfo for dates January 1, 2009, to May 25, 2022. Covidence was used for screening, and the data were extracted following Cochrane's guidelines. The primary end point assessed was the change in the mean plasma concentration of HbA1c at 3 months or more. Cochrane risk of bias 2 was used to assess risk of bias. Data on reach, uptake, and feasibility were summarized narratively, and data on HbA1c reduction were synthesized in a meta-analysis. Grading of Recommendations, Assessment, Development, and Evaluation criteria was used to evaluate the level of evidence.
Results
The search identified 3236 records, from which 56 RCTs from 24 regions (n = 11,486 participants) were included in the narrative synthesis, and 26 studies (n = 4546 participants) in the meta-analysis. Text messages (SMS) were used in 20 studies as the primary mode of delivery of the digital health intervention, 25 used smartphone applications, and 11 used website interventions. The smartphone application interventions reported higher reach compared with the others, but the website-based interventions reported higher uptake compared with the others. Effective interventions, in general, included people with greater severity of their condition at baseline (i.e., higher HbA1c), and the intervention involved a higher intensity—such as more frequent use of a smartphone application. Overall, the digital health intervention group participants had a −0.30 (95% CI, −0.42 to −0.19) percentage point greater reduction in HbA1c, compared with the control group participants. The difference in HbA1c reduction between groups was statistically significant when intervention was delivered through a smartphone application (−0.42% [95% CI, −0.63 to −0.20]) or by text message (−0.37% [95% CI, −0.57 to −0.17]), but not when delivered via a website (−0.09% [95% CI, −0.64 to 0.46]). The level of evidence was moderate overall owing to the considerable heterogeneity of the included studies.
Conclusions
Smartphone apps and text messaging (SMS)—but not website-based interventions—were associated with better glycemic control, but the studies' heterogeneity should be recognized. Considering that both smartphone app and SMS interventions are effective for diabetes management, clinicians should consider factors such as reach, uptake, patient preference, and context of the intervention when deciding on the mode of delivery of the intervention. Clinicians should familiarize themselves with these modalities of program delivery and encourage people with T2D to use evidence-based applications to improve their self-management of diabetes. Future research needs to describe in detail the factors involved in effective implementation of SMS and smartphone application interventions, such as the optimal dose, frequency, timing, user interface, and communication mode.
Comments
This systematic review and meta-analysis included a large number of randomized controlled trials and found that SMS and smartphone-based interventions (but not website-based interventions) improved hemoglobin A1c (HbA1c). The authors also considered implementation outcomes such as reach and uptake. This review introduces the concept that not all digital health interventions are created equally, and that mode of delivery may be as important a consideration as the content of the intervention. Buried in the discussion is a consideration of the impact that hardware and operating system (e.g., iPhone vs android) can have on the time it takes to complete health-related tasks, suggesting that digital health solutions should be considered in the context of the hardware through which they are delivered.
Costs and Cost-effectiveness of Implementing a Digital Diabetes Prevention Program in a Large, Integrated Health System
Smith DH, O'Keeffe-Rosetti M, Fitzpatrick SL, Mayhew M, Firemark AJ, Gruß I, Nyongesa DB, Smith N, Dickerson JF, Stevens VJ, Vollmer WM, Fortmann SP Kaiser Permanente Center for Health Research, Portland, OR
Now that the Diabetes Prevention Program (DPP) has been translated into digital formats, this report provides an economic evaluation of a digital DPP implemented in a large, integrated health-care system.
Methods
Electronic medical record data were used to assess 4148 patients who were invited to participate in digital DPP based on their clinical characteristics (glycated hemoglobin [HbA1c] level 5.7%–6.4% and body mass index ≥ 30 kg/m2). Using a propensity score we matched (1:1) the enrolled patients with nonenrolled patients for a total of 784. We identified high-risk patients (i.e., above the 50th percentile of risk; n = 202) by calculating each patient's 2-year chance of developing diabetes. The cost of the intervention, the costs of medical care over the 12- and 24-month follow-up period, and the incremental cost-effectiveness ratio as the cost per additional kilogram of weight loss at 24 months were determined.
Results
At 12 months, the enrolled patients had lower total costs (8783 [95% CI, 5681–7538]). This pattern attenuated slightly at 24 months (enrolled 18,846 [95% CI, 14,097–16,688]). We found an incremental cost-effectiveness ratio of 150 per additional kilogram of weight loss; at the same willingness-to-pay, there was a 60% chance in the high-risk subgroup. The study's limitations include its nonrandomized design and potential volunteer bias.
Conclusions
Compared with other lifestyle interventions, digital DPP had a favorable cost-effectiveness profile.
Comments
The biggest problem with the original Diabetes Prevention Program is that of scalability—and therefore reach. Digital DPPs can help solve those issues. Omada Health's digitally translated DPP program was previously found to be effective; the present analysis also demonstrates its cost-effectiveness. The authors also address in the discussion the concept that cost efficiency for this and presumably similar programs increases if one focuses on enrolling the highest risk individuals in a clinical cohort.
The Aim2Be mHealth Intervention for Children with Overweight or Obesity and Their Parents: Person-centered Analyses to Uncover Digital Phenotypes
De-Jongh González O1, Tugault-Lafleur CN2, Buckler EJ3, Hamilton J4, Ho J5, Buchholz A6, Morrison KM7, Ball GD8, Mâsse LC1
1School of Population and Public Health, University of British Columbia, BC Children's Hospital Research Institute, Vancouver, BC, Canada; Faculty of Medicine & Dentistry, University of Alberta, Edmonton, AB, Canada; 2School of Nutrition Sciences, Faculty of Health Sciences, The University of Ottawa, Ottawa, ON, Canada; Faculty of Medicine & Dentistry, University of Alberta, Edmonton, AB, Canada; 5School of Exercise Science, Physical and Health Education, University of Victoria, Victoria, BC, Canada; Faculty of Medicine & Dentistry, University of Alberta, Edmonton, AB, Canada; 4Department of Paediatrics, Hospital for Sick Children, University of Toronto, Toronto, ON, Canada; Faculty of Medicine & Dentistry, University of Alberta, Edmonton, AB, Canada; 5Cumming School of Medicine, Department of Pediatrics, University of Calgary, Calgary, AB, Canada; Faculty of Medicine & Dentistry, University of Alberta, Edmonton, AB, Canada; 6Children's Hospital of Eastern Ontario Research Institute, Ottawa, ON, Canada; Faculty of Medicine & Dentistry, University of Alberta, Edmonton, AB, Canada; 7Department of Pediatrics, Center for Metabolism, Obesity and Diabetes Research, McMaster University, Hamilton, ON, Canada; Faculty of Medicine & Dentistry, University of Alberta, Edmonton, AB, Canada; 8Department of Pediatrics, Faculty of Medicine & Dentistry, University of Alberta, Edmonton, AB, Canada
Few studies have characterized user typologies derived from individuals' patterns of interactions with specific app features (digital phenotypes) in mobile health (mHealth) interventions targeting childhood obesity. This study identified digital phenotypes among 214 parent-child dyads who used the Aim2Be mHealth app as part of a randomized controlled trial conducted between 2019 and 2020, and explored whether participants' characteristics and health outcomes differed across phenotypes.
Methods
Distinct parent and child phenotypes were identified using latent class analysis based on their use of the app's features (behavioral, gamified, and social) over 3 months. Multinomial logistic regression models were used to assess whether the phenotypes differed by demographic characteristics. Covariate-adjusted mixed-effect models evaluated changes in BMI z scores (zBMI), diet, physical activity, and screen time across phenotypes.
Results
Five digital phenotypes were identified in the parent: socially engaged (35 of 214, 16.3%) and independently engaged (18 of 214, 8.4%), who used mainly the social or the behavioral features of the app, respectively; fully engaged (26 of 214, 12.1%); partially engaged (32 of 214, 15%); and unengaged (103 of 214, 48.1%) users. The married parents were more likely to be fully engaged than independently engaged (P = 0.02) or unengaged (P = 0.01). The socially engaged parents were older than the fully engaged (P = 0.02) and unengaged (P = 0.01). For the children, the latent class analysis revealed four phenotypes: fully engaged (32 of 214, 15%), partially engaged (61 of 214, 28.5%), dabblers (42 of 214, 19.6%), and unengaged (79 of 214, 36.9%). The fully engaged children were younger than the dabblers (P = 0.04) and unengaged (P = 0.003). The dabblers lived in higher-income households than fully and partially engaged (P = 0.03 and P = 0.047, respectively). Fully engaged children were more likely to have fully engaged (P < 0.001) and partially engaged (P < 0.001) parents than the unengaged children. Compared with unengaged children, the fully and partially engaged children had decreased total sugar (P = 0.006 and P = 0.004, respectively) and energy intake (P = 0.03 and P = 0.04, respectively) after 3 months of app use. The partially engaged children also had decreased sugary beverage intake compared with the unengaged children (P = 0.03). Over time, children with fully engaged parents had decreased zBMI; children with unengaged parents had increased zBMI (P = 0.005). Finally, children with independently engaged parents had decreased caloric intake compared with children with unengaged parents who had increased caloric intake over time (P = 0.02).
Conclusions
The success of mHealth interventions depends on full parent-child engagement. More research is needed into program design elements that can affect participants' engagement in supporting behavior change.
Comments
This article uses an interesting approach: latent class analyses to classify individuals based on their usage patterns of different features of the application. De-Jongh González and colleagues then identified the individual characteristics in both parent and child users that differed between the latent phenotypes. The study suggests that active parental engagement in mHealth interventions may be required to achieve a positive benefit for the child, that each mHealth intervention must have “active ingredients” that promote health behavior change, and that dose–response analysis by feature (rather than relying on overall app engagement metrics) may help to identify patterns in the use of various app features to elicit positive health behaviors. The digital phenotype concept may be critical to accelerating the design of successful mHealth applications.
Effects of E-health-based Interventions on Glycemic Control for Patients with Type 2 Diabetes: A Bayesian Network Meta-analysis
Zhang X, Zhang L, Lin Y, Liu Y, Yang X, Cao W, Ji Y, Chang C
Department of Social Medicine and Health Education, School of Public Health, Peking University Health Science Center, Beijing, China
The deep integration of the internet and health care has led to electronic tools and information technology devoted to disease management in type 2 diabetes (T2D). This study evaluated the effectiveness of different forms and durations of E-health interventions in achieving glycemic control in T2D patients.
Methods
PubMed, Embase, Cochrane, and Clinical Trials databases were searched for randomized controlled trials reporting different forms of E-health interventions for glycemic control in T2D patients, including comprehensive measures, smartphone applications, phone calls, short message service (SMS texts), websites, wearable devices, and usual care. The inclusion criteria were adults aged ≥ 18 with T2D, intervention period of ≥ 1 month, glycated hemoglobin (HbA1c) (%) measured as outcome, and randomized control of E-health–based approaches. Cochrane tools were used to assess the risk of bias and R 4.1.2 to conduct the Bayesian network meta-analysis.
Results
The analysis included a total of 88 studies comprising 13,972 T2D patients. Compared with the usual care group, the SMS-based intervention was superior in reducing HbA1c levels (mean difference, −0.56 [95% CI, −0.82 to −0.31), followed by smartphone apps (mean difference, −0.45 [95% CI, −0.61 to −0.30]), comprehensive measures (mean difference, −0.41 [95% CI, −0.57 to −0.25]), websites (mean difference, −0.39 [95% CI, −0.60 to −0.18]), and phone calls (mean difference, −0.32 [95% CI, −0.50 to −0.14]) (P < 0.05). The most effective interventions lasted ≤ 6 months.
Conclusions
In patients with T2D, all types of E-health–based approaches can improve glycemic control. The most effective intervention, SMS texting, is a high-frequency, low-barrier technology that achieves the best effect in lowering HbA1c, with ≤ 6 months as the optimal intervention duration.
Comments
Zhang and colleagues produced an impressive work that reviews 88 randomized controlled trials that collectively studied e-health interventions in > 13,000 individuals with T2D. The authors identified that SMS-based and smartphone app–based interventions were the most successful in improving disease outcomes. This study essentially suggests that the success of an E-health intervention not only depends on psychological theory or the type of multimedia content, but also on the vehicle of delivery. My question is this: Given this information, why have we not seen more interventions that rely on both a smartphone application and text messaging channels? Would we see even greater efficacy with a combined strategy?
New Digital Health Technologies for Insulin Initiation and Optimization for People with Type 2 Diabetes
Kerr D1, Edelman S2, Vespasiani G3, Khunti K4
1Sansum Diabetes Research Institute, Santa Barbara, CA; University of Leicester, Leicester General Hospital, Leicester, UK; 2University of California San Diego Veterans Affairs Medical Center, San Diego, CA; University of Leicester, Leicester General Hospital, Leicester, UK; 3METEDA S.r.l., Rome, Italy; University of Leicester, Leicester General Hospital, Leicester, UK; 4Diabetes Research Centre, University of Leicester, Leicester General Hospital, Leicester, UK
Many people with type 2 diabetes (T2D) eventually need insulin to help reduce their risk of serious associated complications, but the barriers to the initiation and/or optimization of insulin may expose them to sustained hyperglycemia. New and future technologies may provide opportunities to help overcome these barriers.
Methods
Software tools and devices developed to support the initiation and/or optimization of insulin were identified via a focused literature search of PubMed and key scientific congresses then manually filtering > 300 publications and conference abstracts.
Results
Most new devices to support insulin therapy help track the dose and timing of insulin, and most of the available software tools have been developed for smartphones. Current published data suggest that the use of these technologies is associated with equivalent or improved glycemic outcomes compared with standard care; among the benefits are reduced time burden and improved knowledge of diabetes among health-care providers. Good-quality evidence remains in short supply, however.
Conclusions
New digital health tools may help to reduce barriers to optimal insulin therapy. An integrated solution that connects glucose monitoring, dose recording, and titration advice as well as records comorbidities and lifestyle factors has the potential to reduce the complexity and burden of treatment and may improve adherence to titration and treatment, resulting in better outcomes for people with diabetes.
Comments
Let's face it: health-care providers have many pressures on their time. Current models of health-care delivery do not support closely supervised titration of medication doses in the home setting, and persons with diabetes often cannot get the access they need to the frequent clinical appointments needed to support successful office-based titration. Semiautomation of the titration process using regulatory-approved, mobile app–based algorithms is the answer. So why aren't more clinicians using them? And why aren't more individuals with diabetes who are starting basal insulin asking for them? It may be that these tools must be packaged into a broader digital platform that provides other compelling reasons for the clinician and the individual with diabetes to interact with the platform.
Economic Evaluation Associated with Clinical-grade Mobile App-based Digital Therapeutic Interventions: Systematic Review
Sapanel Y1, Tadeo X1,2, Brenna CTA3, Remus A1,2,4,5, Koerber F6,7, Cloutier LM8, Tremblay G9, Blasiak A1,2,4,10, Hardesty CL11, Yoong J12,13, Ho D1,2,4,10
1The Institute for Digital Medicine WisDM, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 2The N.1 Institute for Health, National University of Singapore, Singapore, Singapore; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 3Department of Anesthesiology & Pain Medicine, University of Toronto, Toronto, ON, Canada; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 4Department of Biomedical Engineering, College of Design and Engineering, National University of Singapore, Singapore, Singapore; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 5Heat Resilience and Performance Centre, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 6IU Internationale Hochschule GmbH, Bad Honnef, Germany; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 7Flying Health GmbH, Berlin, Germany; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 8Department of Analytics, Operations, and Information Technologies, University of Quebec at Montreal, Montreal, QC, Canada; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 9Cytel Canada Health Inc, Toronto, ON, Canada; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 10Department of Pharmacology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 11Pureland Global Venture Pte Ltd, Singapore; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 12Research For Impact, Singapore; Yong Loo Lin School of Medicine, National University of Singapore, Singapore; 13Behavioural and Implementation Science Interventions, Yong Loo Lin School of Medicine, National University of Singapore, Singapore
The adoption and implementation of digital therapeutics (DTx), a class of software-based clinical interventions to prevent, manage, or treat medical conditions as well as deliver portability for patients and scalability for health-care providers, were accelerated by the need for remote care during the COVID-19 pandemic. Awareness about their utility has rapidly grown among providers, payers, and regulators, yet relatively little is known about their capacity to provide economic value in care. This study systematically reviewed and summarized the published evidence regarding the cost-effectiveness of clinical-grade mobile app-based DTx and explored the factors affecting such evaluations.
Methods
A systematic review of economic evaluations of clinical-grade mobile app–based DTx was conducted following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) 2020 guidelines. PubMed, Cochrane Library, and Web of Science and other major electronic databases were searched for eligible studies published from inception to October 28, 2022. Two independent reviewers evaluated the eligibility of all the retrieved articles for inclusion in the review, and methodological quality and risk of bias were assessed for each included study.
Results
A total of 18 studies were included in this review. Of the 18 studies, 7 (39%) were nonrandomized study-based economic evaluations, 6 (33%) were model-based evaluations, and 5 (28%) were randomized clinical trial-based evaluations. The DTx intervention subject to assessment was found to be cost-effective in 12 (67%) studies, cost saving in 5 (28%) studies, and cost-effective in 1 (6%) study in only 1 of the 3 countries where it was being deployed in the final study. Several included studies had qualitative deficiencies in methodology and substantial potential for bias, including risks of performance bias and selection bias in participant recruitment.
Conclusions
DTx interventions offer potential economic benefits, but the economic analyses conducted to date have important methodological shortcomings that must be addressed in future evaluations to reduce the uncertainty surrounding the widespread adoption of DTx interventions.
Comments
It is easy to forget that digital therapeutics (DTx) has only matured in the marketplace since 2017, when the first regulatory approval for a DTx occurred in the United States. The proliferation of fast-track approaches to regulatory approval and integration into reimbursement markets followed. This review evaluates DTx for diabetes through the lens of economics. The reviewers evaluated 7395 records to identify the 18 studies included in the current review. Only two of the studies evaluated DTx for diabetes, but the review is worthy of our attention because of its detailed analysis of the economic value of DTx. We anticipate that the number of DTx in diabetes will grow exponentially, making it essential that we see more high-quality economic evaluations.
Efficacy of Personalized Diabetes Self-care Using an Electronic Medical Record–Integrated Mobile App in Patients with Type 2 Diabetes: 6-Month Randomized Controlled Trial
Lee EY1, Cha SA2, Yun JS3, Lim SY4, Lee JH4, Ahn YB5, Yoon KH1, Hyun MK5, Ko SH3
1Division of Endocrinology and Metabolism, Department of Internal Medicine, Seoul St. Mary's Hospital, College of Medicine, The Catholic University of Korea, Seoul, Republic of Korea; College of Korean Medicine, Dongguk University, Gyeongju, Republic of Korea; 2Division of Endocrinology and Metabolism, Department of Internal Medicine, Wonkwang University Sanbon Hospital, Gunpo, Republic of Korea; College of Korean Medicine, Dongguk University, Gyeongju, Republic of Korea; 3Division of Endocrinology and Metabolism, Department of Internal Medicine, St. Vincent's Hospital, College of Medicine, The Catholic University of Korea, Suwon, Republic of Korea; College of Korean Medicine, Dongguk University, Gyeongju, Republic of Korea; 4Catholic Institute of Smart Healthcare Center, The Catholic University of Korea, Seoul, Republic of Korea; College of Korean Medicine, Dongguk University, Gyeongju, Republic of Korea; 5Department of Preventive Medicine, College of Korean Medicine, Dongguk University, Gyeongju, Republic of Korea
Systems that combine technology and web-based coaching can help treat chronic conditions such as diabetes, but the effectiveness of apps in mobile health (mHealth) interventions has been inconclusive. Heterogeneous interventions, varying follow-up durations, and little randomized controlled trial data and lack of long-term follow-up have all presented difficulties, especially for apps integrated into electronic medical records. This study assessed an electronic medical record–integrated mobile app for personalized self-care in patients with type 2 diabetes mellitus (T2D), focusing on the self-monitoring of blood glucose, glycemic control, and lifestyle modifications.
Methods
A 26-week, three-arm, randomized, controlled, open-label, parallel group trial recruited patients with T2D and a hemoglobin A1c (HbA1c) level of ≥ 7.5%. The mHealth intervention (iCareD system) consisted of self-monitoring of blood glucose with the automatic transfer of glucose, diet, and physical activity counseling data. Participants were randomly assigned to the following three groups: usual care, mobile diabetes self-care (MC), and MC with personalized, bidirectional feedback from physicians (MPC). The primary outcome was the change in HbA1c levels at 26 weeks. After the intervention, diabetes-related self-efficacy, self-care activities, and satisfaction with the iCareD system were assessed.
Results
A total of 269 participants were enrolled, and 234 patients (86.9%) remained in the study at 26 weeks. At 12 weeks after the intervention, the mean decline in HbA1c levels was significantly different among the three groups (usual care vs MC vs MPC: −0.49% vs −0.86% vs −1.04%; P = 0.02). The HbA1c level decreased in all groups, but it did not differ among groups after 26 weeks. In a subgroup analysis, HbA1c levels showed a statistically significant decrease after the intervention in the MPC group compared with the change in the usual care or MC group, especially in patients aged < 65 years (P = 0.02), patients with a diabetes duration of ≥ 10 years (P = 0.02), patients with a body mass index of ≥ 25.0 kg/m2 (P = 0.004), patients with a C-peptide level of ≥ 0.6 ng/mL (P = 0.008), and patients who did not undergo treatment with insulin (P = 0.004) at 12 weeks. A total of 87.2% (137 of 157) of the participants were satisfied with the iCareD system.
Conclusions
The mHealth intervention for diabetes self-care showed short-term efficacy in glycemic control, but the effect decreased over time. The participants were comfortable with using the iCareD system and exhibited high adherence.
Comments
The idea of a mobile app connected to the electronic health record (EHR) to support diabetes self-management and clinician feedback is not new. Multiple such platforms are available globally, although the pace at which they have achieved integration into the EHR has varied by platform and by market. These authors found that the iCareD platform achieved short-term benefit in glycemic control, but this result was not sustainable. This approach is essentially what the United States and the European Union countries have called remote patient monitoring or remote monitoring. This provides further evidence of the potential benefits as well as the limitations of such an approach if one does not design for continued user engagement with the platform across longer time horizons.
Randomized, Controlled Trial of a Digital Behavioral Therapeutic Application to Improve Glycemic Control in Adults with Type 2 Diabetes
Hsia J1,2, Guthrie NL3, Lupinacci P3, Gubbi A1, Denham D4, Berman MA3, Bonaca MP1,2
1CPC Clinical Research, Aurora, CO; San Antonio, TX; 2University of Colorado School of Medicine, Aurora, CO; San Antonio, TX; 3Better Therapeutics, Inc., San Francisco, CA; San Antonio, TX; 4Clinical Trials of Texas, San Antonio, TX
Cognitive behavioral therapy (CBT) is an established form of psychological treatment that endeavors to identify and change unhelpful thinking patterns. This study evaluated the efficacy and safety of a digital therapeutic application (app) delivering CBT designed to improve glycemic control in patients with type 2 diabetes (T2D).
Methods
Adults with T2D and an hemoglobin A1c (HbA1c) of 7% to < 11% were randomly assigned to use a digital therapeutic app delivering CBT (BT-001) or a control app, both in addition to the standard care. The primary study end point was treatment group difference in mean HbA1c change from baseline to 90 days.
Results
Among 669 randomly assigned participants who completed app onboarding, the mean age was 58 years, body mass index of 35 kg/m2, 54% were female, 28% Black, and 16% Latino. The baseline HbA1c was 8.2% and 8.1% in the BT-001 and control groups, respectively. After 90 days of app access, the change in HbA1c was −0.28% (95% CI, −0.41 to −0.15) in the BT-001 group and + 0.11% (95% CI, −0.02 to 0.23) in the control group (treatment group difference 0.39%; P < 0.0001). The HbA1c reduction paralleled the participants' exposure to the therapeutic intervention, which was assessed as the number of modules completed on the app (P for trend < 0.0001). No adverse events in either group were attributed to app use, and no adverse device effects were reported.
Conclusions
Patients randomly assigned to the BT-001 arm relative to the control arm had significantly lower HbA1c levels at 90 days. The digital therapeutic may provide a scalable treatment option for patients with T2D.
Comments
The authors here demonstrate the value of incorporating evidence-based psychological technologies into a digital therapeutic or mHealth application. Specifically, cognitive behavioral therapy addressing the participants' core knowledge, attitudes, and beliefs related to different lifestyle and self-management topics demonstrated efficacy for reducing HbA1c over 90 days in a dose-dependent manner. More digital therapeutics should consider incorporating psychological technologies to boost efficacy for improving the targeted health outcome.
Economic Evaluations of mHealth Interventions for the Management of Type 2 Diabetes: A Scoping Review
Tornvall I1,2, Kenny D1, Wubishet BL3, Russell A1,4,5, Menon A1,6, Comans T1
1Centre for Health Services Research, The University of Queensland, Brisbane, QLD, Australia; Metro South Health, Brisbane, QLD, Australia; 2Centre for Health Services Research, Faculty of Medicine, The University of Queensland, Brisbane, QLD, Australia; Metro South Health, Brisbane, QLD, Australia; 3Centre for Economic Impacts of Genomic Medicine, Macquarie Business School, Macquarie University, Sydney, NSW, Australia; Metro South Health, Brisbane, QLD, Australia; 4Department of Endocrinology, Alfred Health, Melbourne, VIC, Australia; Metro South Health, Brisbane, QLD, Australia; 5School of Public and Preventive Health, Monash University, Melbourne, VIC, Australia; Metro South Health, Brisbane, QLD, Australia; 6Department of Endocrinology, Metro South Health, Brisbane, QLD, Australia
Research into the clinical benefits of mHealth interventions for type 2 diabetes (T2D) has been limited, and there has yet to be support for the claims of cost-effectiveness or cost savings. This review summarized and critically analyzed the current body of economic evaluation (EE) studies for mHealth interventions for T2D.
Methods
Five databases were comprehensively searched for full and partial EE studies for mHealth interventions for T2D from January 2007 to March 2022. The analysis defined “mHealth” as any intervention that used a mobile device with cellular technology to collect and/or provide data or information for the management of T2D. The CHEERS 2022 checklist was used to appraise the reporting of the full EEs.
Results
The review comprised 12 studies: nine full and three partial evaluations. Text messages smartphone applications were the most common mHealth features. The majority of interventions also included a Bluetooth-connected medical device (e.g., glucose or blood pressure monitors). All studies reported their intervention to be cost-effective or cost-saving, but their reports were of moderate quality, with a median CHEERS score of 59%.
Conclusions
The quality of the reporting on the economics of mHealth interventions for T2D must be substantially improved. Heterogeneity makes it difficult to compare their outcomes, and the failure to report on key items leaves insufficient information for decision-makers to consider.
Comments
The review by Tornvall and colleagues (N = 12 studies) supports the claim that mHealth interventions can be cost-saving or cost-neutral, but the article recommends improving the quality of reporting of economic evaluations. They suggest that authors stop assuming that the cost of a digital health intervention is zero, as this is almost never the case. They also argue for increased use of the CHEERS 2022 checklist for evaluations when writing about the economics of digital therapeutics.
Effectiveness of a Smartphone App (MINISTOP 2.0) Integrated in Primary Child Health Care to Promote Healthy Diet and Physical Activity Behaviors and Prevent Obesity in Preschool-aged Children: Randomized Controlled Trial
Alexandrou C1,2, Henriksson H1, Henström M2, Henriksson P1, Delisle Nyström C2, Bendtsen M1, Löf M1,2
1Department of Health, Medicine and Caring Sciences, Division of Society and Health, Linköping University, Linköping, Sweden; Karolinska Institutet, NEO, Huddinge, Sweden; 2Department of Biosciences and Nutrition, Karolinska Institutet, NEO, Huddinge, Sweden
Previously the efficacy of a parent-oriented mobile health (mHealth) app-based intervention (MINISTOP 1.0) was reported, which showed improvements in healthy lifestyle behaviors, but the effectiveness of the app in real-world conditions had yet to be established. The real-world effectiveness of a 6-month MINISTOP 2.0 intervention on children's intake of fruits, vegetables, sweet and savory treats, sweet drinks, moderate-to-vigorous physical activity, and screen time (primary outcomes) were evaluated as well as parental self-efficacy (PSE) for promoting healthy lifestyle behaviors and children's body mass index (BMI) (secondary outcomes).
Methods
A hybrid type 1 effectiveness-implementation design was used for a two-arm, individually randomized, controlled trial. Parents (n = 552) of 2.5- to 3-year-old children were recruited from 19 child health-care centers across Sweden and randomized to either standard care (control) or the intervention group (MINISTOP 2.0 app). To increase reach, the app was adapted and translated into English, Somali, and Arabic, and all recruitment and data collection were conducted by nurses. The outcomes were assessed at baseline and after 6 months using standardized measures (body mass index [BMI]) and a questionnaire (health behaviors and PSE).
Results
Among the participating parents (n = 552, age: 34.1 ± 5.0 years), 79% were women, and 62% had a university degree. Twenty-four percent (n = 132) of children had two foreign-born parents. At follow-up, the parents in the intervention group reported lower intakes of sweet and savory treats (−6.97 g/day; P = 0.001), sweet drinks (−31.52 g/day; P < 0.001), and screen time (−7.00 min/day; P = 0.012) in their children compared with the control group. Compared with the control group, the intervention group reported higher total PSE (0.91; P = 0.006), PSE for promoting healthy diet (0.34; P = 0.008), and PSE for promoting physical activity behaviors (0.31; P = 0.009). No statistically significant effect was observed for children's BMI z score. Overall, the parents reported high satisfaction with the app, and 54% reported using the app at least once a week.
Conclusions
In this real-world effectiveness trial of the MINISTOP 2.0 app, parents in the intervention group reported decreased intake of sweets and sweet drinks and less screen time (primary outcomes) in their children. The parents also reported higher PSE for promoting healthy lifestyle behaviors.
Comments
This Swedish trial follows a prior randomized clinical trial by presenting real-world evidence for the MINISTOP intervention in 552 parent-child dyads and incorporates a hybrid effectiveness-implementation design. For digital health interventions, it is essential that innovators consider barriers and facilitators to implementation from the earliest stages. The RE-AIM framework (reach, effectiveness, adoption, implementation, and maintenance) is a popular framework (among 170+ theories, models, and frameworks) for the planning and evaluation of dissemination and implementation initiatives. It would be good to see more randomized trials and real-world evidence studies evaluating implementation outcomes.
A Quantitative Model to Ensure Capacity Sufficient for Timely Access to Care in a Remote Patient Monitoring Program
Chang A1,2, Gao MZ2, Ferstad JO2, Dupenloup P2, Zaharieva DP3, Maahs DM3,4, Prahalad P3, Johari R2,4, Scheinker D2,3,4
1Icahn School of Medicine at Mount Sinai, New York, NY; Stanford University, Stanford, CA; 2Department of Management Science and Engineering, Stanford University, Stanford, CA; Stanford University, Stanford, CA; 3Department of Paediatric, Division of Paediatric Endocrinology, Stanford University, Stanford, CA; Stanford University, Stanford, CA; 4Stanford Diabetes Research Centre, Stanford University, Stanford, CA
No standardized models are available for clinics developing and deploying algorithm-enabled remote patient monitoring (RPM) programs to ensure capacity sufficient for timely access to care. This study presents a flexible model and interactive dashboard of capacity planning for whole-population RPM-based care for type 1 diabetes (T1D).
Methods
A weekly RPM program for 277 pediatric patients with T1D at a pediatric academic medical center provided 2 years of observational operational data. Analysis of these data and iterative interviews with the care team identified the primary operational, population, and workforce metrics that drive demand for care providers. An interactive model was designed based on these metrics to facilitate capacity planning, which was deployed as a dashboard.
Results
The three primary population-level drivers of demand are (1) the number of patients in the program, (2) the rate at which patients enroll and graduate from the program, and (3) the average frequency at which patients require a review of their data. The three primary modifiable clinic-level drivers of capacity are (1) the number of care providers, (2) the time required to review patient data and contact a patient, and (3) the number of hours each provider allocates to the program each week. The model identified a variety of practical operational approaches to better match the demand for patient care at the institution studied.
Conclusions
This generalizable, systematic model for capacity planning was designed for a pediatric endocrinology clinic providing RPM for T1D. Deployed as an interactive dashboard, this model facilitated the expansion of a novel care program (4 T Study) for patients with newly diagnosed T1D, and it may facilitate the systematic design of other RPM-based care programs.
Comments
Chang and colleagues have previously published their observational outcomes on the use of a CGM-focused population health management tool to support risk-based remote patient monitoring (RPM) for new-onset pediatric T1D. One thing we know about implementing new technologies in clinical settings: clinical staff must divert their attention away from the activities where they are currently allocating effort and toward activities that operationalize the new digital health technology or other intervention. These authors introduced a web-based capacity planning tool that serves as a nice companion to implementing their technology—or potentially to any digital health technology implementation.
Remote Glucose Monitoring Is Feasible for Patients and Providers Using a Commercially Available Population Health Platform
Crossen SS1,2, Romero CC1, Lewis C2, Glaser NS1
1Department of Pediatrics, University of California Davis, Sacramento, CA; University of California Davis, Sacramento, CA; 2Center for Healthcare Policy and Research, University of California Davis, Sacramento, CA
Remote patient monitoring (RPM) may enable more individualized and effective care for patients with type 1 diabetes (T1D), but it requires population analytics to focus limited clinical resources on the patients most in need. Using a commercially available data analytic platform (glooko Population Health) among a cohort of youth with T1D, this study explored the feasibility of RPM from patient and provider standpoints.
Methods
The study recruited patients aged 1–20 years with established T1D (≥ 12 months) and CGM use (≥ 3 months) or their caregivers. The participants' CGM devices were connected to the glooko app, which was linked to the research team's glooko account during a 1-month baseline period. During the 6-month intervention period, the participants with > 15% of glucose values > 250 mg/dL or > 5% of values < 70 mg/dL each month were contacted by a pediatric endocrinologist either by text messaging or by telephone (with an interpreter, if needed) for personalized diabetes management recommendations. Afterward the effects on glycemic control were estimated via change in glucose management indicator (GMI) generated from CGM data at baseline and completion, and the participants were surveyed about their experiences. Changes in time spent within various glucose ranges were also evaluated, and all glycemic metrics were compared with a nonrandomized control group via difference-in-difference regression, adjusting for baseline characteristics.
Results
Remote data-sharing was successful for 36 of the 39 participants (92%). Between 33% and 66% of participants merited outreach each month, and clinician outreach efforts required a median of 10 minutes per event. RPM was reported to be helpful by 94% of participants, and was associated with a GMI change of −0.25% (P = 0.047) for the entire cohort. Stratified analysis revealed the greatest treatment effects were found among participants with a baseline GMI of 8.0%–9.4% (GMI change of −0.68%, P = 0.047; 19.84% reduction in time spent > 250 mg/dL, P = 0.005).
Conclusions
RPM for patients with T1D using a commercially available population health platform was not only feasible but was considered helpful by patients/caregivers, particularly those with suboptimal glycemic control who found the clinician-initiated outreach particularly beneficial.
Comments
Crossen and colleagues have demonstrated that a data integration and visualization platform for diabetes technologies (glooko Population Health) meets the definition of software as a medical device and can be effectively leveraged to support RPM in pediatric T1D. This provides the clinic with access to a new type of revenue from RPM-focused Current Procedural Terminology (CPT) codes. In reviewing the literature, we have not found any other pediatric diabetes programs implementing RPM, so this is an exciting new development.
A Model to Design Financially Sustainable Algorithm-enabled Remote Patient Monitoring for Pediatric Type 1 Diabetes Care
Dupenloup P1, Pei RL1, Chang A1, Gao MZ1, Prahalad P2,3, Johari R1,2, Schulman K4,5, Addala A2, Zaharieva DP2, Maahs DM2,3, Scheinker D1,2,4,6 on behalf of the 4T Research Team
1Department of Management Science and Engineering, Stanford University, Stanford, CA; Division of Biomedical Informatics Research, Stanford University, Stanford, CA; 2Department of Pediatrics, Division of Pediatric Endocrinology, Stanford University, Stanford, CA; Division of Biomedical Informatics Research, Stanford University, Stanford, CA; 3Stanford Diabetes Research Center, Stanford University, Stanford, CA; Division of Biomedical Informatics Research, Stanford University, Stanford, CA; 4Clinical Excellence Research Center, Stanford University, Stanford, CA; Division of Biomedical Informatics Research, Stanford University, Stanford, CA; 5Graduate School of Business, Stanford University, Stanford, CA; Division of Biomedical Informatics Research, Stanford University, Stanford, CA; 6Department of Medicine, Division of Biomedical Informatics Research, Stanford University, Stanford, CA
The existing reimbursement models for population-level, algorithm-enabled, remote patient monitoring (RPM) based on continuous glucose monitor (CGM) data review are geared toward direct provision of clinic care, not population health management. This study developed a financial model to assist pediatric type 1 diabetes (T1D) clinics in designing financially sustainable RPM programs based on algorithm-enabled review of CGM data.
Methods
A weekly RPM program for 302 pediatric patients with T1D at Lucile Packard Children's Hospital provided the data. The customizable financial model calculated the yearly marginal costs and revenues of providing diabetes education. The baseline or status quo scenario was compared with two different care-delivery scenarios in which routine appointments are supplemented with algorithm-enabled, flexible, message-based contacts delivered according to patient need. This model can estimate the minimum reimbursement rate needed for telemedicine contacts to maintain revenue-neutrality and not suffer an adverse impact to the bottom line.
Results
In both scenarios, the financial model estimated that an average reimbursement rate of roughly US$10.00 per telehealth interaction would be sufficient to maintain revenue neutrality. Algorithm-enabled RPM could potentially be billed using existing RPM Current Procedural Terminology (CPT) codes and lead to margin expansion.
Conclusions
The model evaluated the financial impact of adopting algorithm-enabled RPM in a pediatric endocrinology clinic serving T1D patients. This model establishes a clear threshold reimbursement value for maintaining revenue-neutrality as well as an estimate of potential RPM reimbursement revenue. This may serve as a useful financial-planning tool for a pediatric T1D clinic seeking to leverage algorithm-enabled RPM to provide flexible, more timely interventions to its patients.
Comments
Dupenloup and colleagues have provided the first model for evaluating the financial impact of implementing risk-based population health management in diabetes care. This important study provides a roadmap for others implementing digital health tools, with an eye toward helping clinicians in an ambulatory setting identify the at-risk patients and to escalate the level of care using RPM. These data should be delivered to hospital administrators at every pediatric (and adult) diabetes center to help centers accelerate their conversion to an RPM-inclusive model of care.
Cost-utility of an Online Education Platform and Diabetes Personal Health Record: Analysis over Ten Years
Cunningham SG1, Stoddart A2, Wild SH3, Conway NJ1, Gray AM4, Wake DJ3
1School of Medicine, University of Dundee, Dundee, UK; Department of Public Health, University of Oxford, Oxford, UK; 2Edinburgh Clinical Trials Unit, Usher Institute, The University of Edinburgh, Edinburgh, UK; Department of Public Health, University of Oxford, Oxford, UK; 3Usher Institute, The University of Edinburgh, Edinburgh, UK; Department of Public Health, University of Oxford, Oxford, UK; 4Health Economics Research Centre, Department of Public Health, University of Oxford, Oxford, UK
Scotland's interactive website and mobile app for people with diabetes and their caregivers, My Diabetes My Way (MDMW), provides multimedia resources for diabetes education and offers access to electronic personal health records. This study assessed the cost-utility of MDMW compared with routine diabetes care in people with type 2 diabetes (T2D) who do not use insulin.
Methods
The clinical parameters of MDMW users (n = 2576) were compared with a matched cohort of individuals receiving only routine care (n = 11 628). The matched criteria were age, diabetes duration, sex, and socioeconomic status. The analysis used the United Kingdom Prospective Diabetes Study (UKPDS) Outcomes Model 2, and examined the impact on life expectancy, quality-adjusted life years (QALYs), and costs of treatment and complications simulated over 10 years, including a 10% sensitivity analysis.
Results
The MDMW cohort had 1670 men (64.8%), average age of 64.3 years, and duration of diabetes of 5.5 years. For the 906 women (35.2%), the average age was 61.6 years, and duration of diabetes was 4.7 years. The cumulative mean QALY gain was 0.054 (95% CI, 0.044–0.062) years. The mean difference in cost was −£118.72 (95% CI, −£150.16 to −£54.16) over 10 years; with increasing MDMW costs (10%): −£50.49 (95% CI, −£82.24 to £14.14), or decreasing MDMW costs (10%): −£186.95 (95% CI, −£218.53 to −£122.51).
Conclusions
MDMW may be among the most cost-effective interventions currently available to support diabetes. MDMW is “dominant” over usual care (cost-saving and life-improving) in supporting self-management in people with T2D who are not treated with insulin. Its wider use may result in significant cost savings through delay or reduction of long-term complications and improved QALYs in Scotland and other countries.
Comments
Cunningham and colleagues describe how education, tracking, and technological linking to an electronic health record can be used to improve outcomes for people with T2D in various geographic areas (in this case Scotland). Outcomes were improved compared with controls. Because enrollment in the intervention was by user choice, the improved outcomes may have been based on selection bias. Nonetheless, the approach makes sense and can be used by other jurisdictions or integrated health-care settings to provide support to patients and their caregivers, which is critically needed.
A Randomised Control Trial to Explore the Impact and Efficacy of the Healum Collaborative Care Planning Software and App on Condition Management in the Type 2 Diabetes Mellitus Population in NHS Primary Care
Heald AH1,2, Roberts S3, Albeda Gimeno L3, Gilingham E, James M4,5, White A5, Saboo A3, Beresford L5, Crofts A6, Abraham J3
1Department of Diabetes and Endocrinology, Salford Royal Hospital, Salford, UK; Southport, UK; 2The School of Medicine, Manchester Academic Health Sciences Centre, University of Manchester, Manchester, UK; Southport, UK; 3Healum Ltd, London, UK; Southport, UK; 4Park Lane Surgery, Waters Green Medical Centre, Macclesfield, UK; Southport, UK; 5Middlewood Partnership, Bollington, Cheshire, UK; Southport, UK; 6Alan Crofts Evaluation, Southport, UK
Personalized care planning is an effective means of improving health outcomes and experiences of care among people with type 2 diabetes (T2D) and other long-term health conditions. This study examines a specific example of one such effective, scalable intervention.
Methods
The study enrolled 197 participants with T2D who were randomized into two groups: 115 participants for the active intervention with digital health planning (app + usual care), and 82 participants receiving the usual care (control group). The outcomes analyzed were changes in body mass index (BMI) and glycated hemoglobin (HbA1c) over the 6-month follow-up period. The participants also received questionnaires, and those in the active treatment group were interviewed to create their care plan and ensure their access to the software application.
Results
The active treatment group had significant reductions in HbA1c (P < 0.01) and BMI (P < 0.037) compared with the control group (no significant change). The average percentage change in HbA1c for the treatment group over 6 months was −7.4% ± 1.4% (SE) compared with 1.8% ± 2.1% SE for the control group. The average percentage change in BMI for the treatment group was −0.7% ± 0.4% SE, compared with −0.2% ± 0.5% SE for the control group. A higher percentage of the active treatment group reduced their HbA1c and BMI than the control group. For HbA1c, 72.4% of the active treatment group reduced their HbA1c level compared with 41.5% in the control group. For BMI, the active treatment group experienced a 52.7% reduction, compared with 42.9% for the control group. Self-measured quality of life improved for the patients in the active treatment group, as shown by an increase in their pretrial to posttrial EQ-5D-5L questionnaire ratings by an average of 0.0464 ± 0.0625 SE, compared with a decrease of 0.0086 ± 0.0530 SE for the control group. The average EQ VAS score also increased on average by 8.2% from pretrial to posttrial for the active treatment group; it decreased by an average of −2.8% for the control group.
Conclusions
The provision of personalized plans of care, support, and education linked to a mobile app can result in HbA1c and BMI reduction for many individuals with T2D. A patient management app as well as a personalized care plan also led to an improvement in patient self-rated quality of life and engagement.
Comments
England's National Health Service (NHS) has been on the forefront of countless innovations over the decades, and this study illustrates another one. Heald and colleagues describe a randomized controlled trial in which a collaborative care planning software and app, plus education and support, improved diabetes management in the experimental group compared with the control group. The study only measured outcomes over 6 months of treatment, but it is hoped that continued use, though not easy to maintain, will sustain the impact.
Evaluating the Implementation of a Digital Diabetes Prevention Program in an Integrated Health Care Delivery System among Older Adults: Results of a Natural Experiment
Fitzpatrick SL1, Mayhew M1, Rawlings AM1, Smith N1, Nyongesa DB1, Vollmer WM1, Stevens VJ1, Grall SK2, Fortmann SP1
1Kaiser Permanente Center for Health Research, Portland, OR; Portland, OR; 2Kaiser Permanente Northwest, Portland, OR
This natural experiment study assessed the effectiveness of a 12-month digital Diabetes Prevention Program (DPP) for adults aged 65–75 years with prediabetes and obesity within a large, integrated health-care system.
Methods
In this 12-month intervention with 12-month follow-up, 472 participants used the DPP (health coaching from lifestyle coaches, virtual small-group support, and electronic behavioral tracking tools for nutrition, physical activity, and weight), and 3432 received standard care. Participants had to be current members of the health plan, with the following measures documented within the previous 12 months: age 65–75 years, hemoglobin A1c (HbA1c) of 39–46 mmol/mol (5.7%–6.4%), and body mass index (BMI) of ≥ 30 kg/m2. Participants could have no prior diagnosis of type 1 or type 2 diabetes (T2D). The primary outcome was change in weight over 12 months from baseline, and secondary outcomes were change in weight over 24 months and change in HbA1c over 12 and 24 months, with data collected from the electronic health record.
Results
Adjusting for propensity scores and covariates, patients in the digital DPP group had a mean weight loss of 8.6 lb (3.9 kg) over 12 months and 5.7 lb (2.6 kg) by 24 months, compared with the steady but minimal weight loss of 1.3 lb (0.59 kg) over 12 months and 2.8 lb (1.3 kg) by 24 months in the control patients. There was a significant difference in the mean change in HbA1c between the DPP group and the control patients over 12 months (−0.10%), but not by 24 months (−0.06%).
Conclusions
Digital DPP interventions appear to be an effective weight loss option and provide potential diabetes prevention for older adults at high risk for T2D.
Comments
Another important study from Kaiser Permanente and digital health company Omada Health has demonstrated that virtual diabetes prevention programs (DPP) work and provide improvements for those who participate. Using a natural experiment and a statistical approach called propensity matching allowed for proof of efficacy without requiring a RCT. This approach is gaining more acceptance in the practical world of digital therapeutics. Traditional RCTs are very expensive and take such a long time to plan, implement, and evaluate that the digital intervention being studied ultimately may be many generations removed from current technologies and programs when the trial is completed. It is great to see the increasing availability of high-quality studies demonstrating the effectiveness of virtual DPP interventions.
The Effectiveness of Digital Health Technologies for Patients with Diabetes Mellitus: A Systematic Review
Stevens S1,2, Gallagher S1, Andrews T1,3, Ashall-Payne L1,3, Humphreys L1, Leigh S1,3
1Research Department, Organisation for the Review of Care and Health Applications, Daresbury, UK; The University of Warwick, Coventry, UK; 2Centre for Health Technology, University of Plymouth, Plymouth, UK; The University of Warwick, Coventry, UK; 3Warwick Medical School (WMS), The University of Warwick, Coventry, UK
Digital health technologies (DHTs), which include mobile health apps (mHealth), have been rapidly gaining popularity for the self-management of chronic diseases such as diabetes, particularly after the COVID-19 pandemic. Although a variety of diabetes-specific mHealth applications exist on the market, the evidence supporting their clinical effectiveness is still limited.
Methods
In this systematic review, a search was conducted of PubMed with additional manual searches of reference lists and Google Scholar to identify randomized controlled trials (RCTs) of mHealth interventions in diabetes published between June 2010 and June 2020. The studies were categorized by the type of diabetes, and the impact of the diabetes-specific mHealth app on the management of glycated hemoglobin (HbA1c) was analyzed.
Results
In a total of 25 studies, comprising 3360 patients, the methodological quality of the trials was mixed. Overall, the participants diagnosed with type 1 diabetes (T1D), T2D, and prediabetes all demonstrated greater improvements in HbA1c as a result of using a DHT compared with those who received the usual care. The analysis revealed an overall improvement in HbA1c compared with usual care, with a mean difference of −0.56% for T1D, −0.90% for T2D, and −0.26% for prediabetes.
Conclusions
DHT can be an effective intervention in the management of HbA1c in diabetes, but further research is needed on the wider clinical effectiveness of diabetes-specific mHealth applications, specifically for T1D and prediabetes and including additional outcomes measures of short-term glycemic variability or hypoglycemic events.
Comments
This literature review identified 25 studies using digital health technologies (DHTs) to improve outcomes for patients with T1D, T2D, and prediabetes. Overall, the participants in each category demonstrated greater improvements in HbA1c because they were using a DHT compared with those who experienced usual care. The review highlights the need for more and better studies and also for reviews that separate the types of diabetes because they are so different.
Weight Loss in a Digital Diabetes Prevention Program for People in Health Professional Shortage and Rural Areas
Graham SA, Auster-Gussman LA, Lockwood KG, Branch OH
Lark Health, Mountain View, CA
Individuals with prediabetes living in hard-to-reach and underserved areas experience barriers to accessing traditional in-person preventive health services. The National Diabetes Prevention Program (DPP) is designed to reduce the risk of developing type 2 diabetes (T2D). Although an increasing number of remote DPPs are available, there are few data on their clinical outcomes for members living in hard-to-reach and underserved areas.
Methods
This study assessed whether living in a designated Health Professional Shortage Area (HPSA) and a rural versus urban area impacted the weight loss of 7266 members of Lark DPP, a fully digital program. The secondary analyses included between-group comparisons of program retention and member characteristics, demographics, and socioeconomics.
Results
The percentage of weight loss did not differ by HPSA (P = 0.16) or rural/urban status (P = 0.15), despite the greater potential barriers for the members residing in HPSAs, who had the highest starting body mass index, lowest incomes, and lowest level of education. The mean percentage of weight loss by members residing in an HPSA + rural area was 4.75% ± 0.09% SE; for members in a non-HPSA + rural area, 4.96% ± 0.16% SE; for members in an HPSA + urban area, 4.55% ± 0.13% SE; and for members in a non-HPSA + urban area, 4.77% ± 0.13% SE. Members who participated in the fully digital DPP achieved weight loss that did not differ by HPSA or urban/rural designation.
Conclusions
Fully digital programs offer a solution to reduce the risk of T2D in areas where the residents may not otherwise have access to diabetes prevention services.
Comments
This study attempted to determine whether the impact of a digital Diabetes Prevention Program (DPP) was different for the users who lived in Health Professional Shortage Areas (HPSA) and rural areas. Despite the greater potential barriers for members residing in HPSAs and rural areas, the digital program was equally effective across the population types. These results are exciting to see, given the bias that low-income populations may not be able to use digital interventions—that may have been the case in the early years of digital therapeutics (pre-2010 or so), but modern interventions have come a long way toward making them effective in a variety of audiences.
Usage and Daily Attrition of a Smartphone-based Health Behavior Intervention: Randomized Controlled Trial
Egilsson E1, Bjarnason R2,3, Njardvik U1
1Department of Psychology, University of Iceland, Reykjavik, Iceland; University of Iceland, Reykjavik, Iceland; 2Department of Pediatrics, University of Iceland, Reykjavik, Iceland; University of Iceland, Reykjavik, Iceland; 3Faculty of Medicine, University of Iceland, Reykjavik, Iceland
Adolescents with smartphones seldom use them for mobile health (mHealth) apps for health improvement, and adolescent mHealth interventions have high attrition rates. Research on these interventions among adolescents has frequently lacked detailed time-related attrition data alongside analysis of attrition reasons through usage. This study analyzed app usage data to obtain daily attrition rates among adolescents in an mHealth intervention to gain a deeper understanding of attrition patterns, including the role of motivational support such as altruistic rewards.
Methods
This randomized controlled trial recruited 304 adolescent participants (152 boys and 152 girls) aged 13 to 15 years. Based on the three participating schools, the participants were randomly assigned to control, treatment as usual (TAU), and intervention groups. Measures were obtained at baseline, continuously throughout the 42-day trial period (research groups), and at the trial end. The mHealth app (SidekickHealth) was a social health game with three main categories: nutrition, mental health, and physical health. The primary measures were attrition based on time from launch, and the type, frequency, and time of health behavior exercise usage. Outcome differences were obtained through comparison tests, and regression models and survival analyses were used for attrition measures.
Results
Attrition differed significantly between the intervention and TAU groups (44.4% vs 94.3%; χ21 = 61.220; P < 0.001). The mean usage duration was 6.286 days in the TAU group and 24.975 days in the intervention group. In the intervention group, the male participants were active significantly longer than the female participants (29.155 vs 20.433 days; χ21 = 6.574; P < 0.001). In the intervention group, the participants completed a larger number of health exercises in all trial weeks; in the TAU group, a significant decrease in usage was observed from the first to second week (t105 = 9.208; P < 0.001) that was not observed in the intervention group. A significant increase in health exercises in the intervention group was seen from the fifth to sixth weeks (t105 = 3.446; P < 0.001) that was not evident in the TAU group. The research group was significantly related to attrition time (hazard ratio, 0.308 [95% CI, 0.222–0.420]) as well as the numbers of mental health exercises (P < 0.001) and nutrition exercises (P < 0.001).
Conclusions
In adolescent mHealth interventions, motivational support is a significant factor for lowering attrition. The results point to sensitivity periods in the completion of diverse health tasks; time-specific attrition, along with the type, frequency, and time of health behavior exercise usage, should be further explored.
Comments
Egilsson and colleagues have highlighted the impact of a smartphone-based, therapeutic game designed to increase the healthy behaviors of adolescents aged 13–15 years old. The analysis focused on daily attrition rate and found that the participants who were randomly assigned to the gamified mHealth app were more engaged than those receiving treatment as usual. (Go figure!) So if you offer adolescents a phone-based experience, they will use it. It is hoped that this approach will lead to actual sustainable changes in behaviors and improved health outcomes.
Digital Health Technologies Enabling Partnerships in Chronic Care Management: Scoping Review
Wannheden C1, Åberg-Wennerholm1 M, Dahlberg M1, Revenäs Å1,2,3, Tolf S1, Eftimovska E1, Brommels M1
1Medical Management Centre, Department of Learning, Informatics, Management and Ethics, Karolinska Institutet, Stockholm, Sweden; County of Västmanland, Uppsala University, Västerås, Sweden; 2Division of Physiotherapy, School of Health Care and Social Welfare, Mälardalen University, Västerås, Sweden; County of Västmanland, Uppsala University, Västerås, Sweden; 3Center for Clinical Research, County of Västmanland, Uppsala University, Västerås, Sweden
Because an increasing number of patients expect to play a greater role in their treatment and care decisions, collaborative health-care practices among interprofessional health-care teams and patients, their families, caregivers, and communities are needed. In recent years, participatory health technologies—digital health technologies that support self-care and collaboration between the community and health-care providers—have received increasing attention, but there is limited knowledge about what features of such technologies should have to support effective patient–professional partnerships. This review mapped and assessed published studies on participatory health technologies supporting partnerships among patients, caregivers, and health-care professionals in chronic care, with a specific focus on identifying these technologies' main features.
Methods
This scoping review of scientific publications in English between January 2008 and December 2020 searched the PubMed and Web of Science databases for peer-reviewed qualitative, quantitative, and mixed methods studies that evaluated digital health technologies for patient-professional partnerships in chronic care settings. The data were charted and analyzed thematically. The PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews) checklist was used.
Results
This review included 32 studies, reported in 34 articles. Most of the articles originated from the United States and Norway, and the topic of participatory health technologies experienced a slightly increasing trend across publication years. Diabetes and cardiovascular diseases were the most common conditions addressed. Of the 32 studies, 12 (38%) evaluated the influence of participatory health technologies on partnerships. The six common features of participatory health technologies were patient–professional communication, self-monitoring, tailored self-care support, self-care education, care planning, and community forums for peer-to-peer interactions. Mostly the outcomes were positive, but it was evident that partnership relationships and the nature of collaborative work could be challenged when the roles and expectations among users were unclear.
Conclusions
Clarifying mutual expectations and carefully considering the implications that the introduction of participatory health technologies may have on the work of patients and health-care professionals, both individually and in collaboration, are important factors. A knowledge gap remains in this area regarding the use of participatory health technologies to effectively support patient–professional partnerships in chronic care management.
Comments
Are health-care providers able to partner with their patients? This scoping review, which identified a variety of studies on this topic, mostly from the United States and Norway, highlighted the key elements required for effective partnerships in chronic care management. The six common features they identified of participatory health technologies were patient–professional communication, self-monitoring, tailored self-care support, self-care education, care planning, and community forums for peer-to-peer interactions. This review provides important information for those who create and implement such studies.
Effects of Mobile Health Interventions on Health-related Outcomes in Older Adults with Type 2 Diabetes: A Systematic Review and Meta-analysis
Lee JJN1, Abdul Aziz A2, Chan ST2, Raja Abdul Sahrizan RSFB2, Ooi AYY2, Teh YT2, Iqbal U3,4, Ismail NA5, Yang A6, Yang J7,8, Teh DBL1,9,10,11, Lim LL2,6,12
1Bia-Echo Asia Center for Reproductive Longevity & Equality (ACRLE), Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Hong Kong, China; 2Department of Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia; Hong Kong, China; 3Global Health & Health Security Department, College of Public Health, Taipei Medical University, Taipei, Taiwan; Hong Kong, China; 4Health ICT, Department of Health, Tasmania, Australia; Hong Kong, China; 5Department of Economics and Applied Statistics, Faculty of Business and Economics, University of Malaya, Kuala Lumpur, Malaysia; Hong Kong, China; 6Department of Medicine and Therapeutics, The Chinese University of Hong Kong, Hong Kong, China; Hong Kong, China; 7College of Earth and Environmental Sciences, Lanzhou University, Lanzhou, China; Hong Kong, China; 8School of Public Health and Social Work, Queensland University of Technology, Brisbane, Queensland, Australia; Hong Kong, China; 9Department of Ophthalmology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Hong Kong, China; 10Department of Anatomy, Yong Loo Lin School of Medicine, National University of Singapore, Singapore; Hong Kong, China; 11Neurobiology Programme, Life Science Institute, National University of Singapore, Singapore; Hong Kong, China; 12Asia Diabetes Foundation, Hong Kong, China
The efficacy of mobile health (mHealth) interventions for improving patients' self-management among the elderly with type 2 diabetes (T2D) has not been explored. The present review examined the effectiveness of mHealth interventions on cardiometabolic outcomes in older adults with T2D.
Methods
After a systematic search from the inception until May 31, 2021, in the MEDLINE, Embase, and PubMed databases, 16 randomized controlled trials were included in the analysis.
Results
The results showed significant improvements in glycosylated hemoglobin (HbA1c) (mean difference −0.24% [95% CI, −0.44 to −0.05], P = 0.01), postprandial blood glucose (−2.91 mmol/L [95% CI, −4.78 to −1.03], P = 0.002), and triglycerides (−0.09 mmol/L [95% CI, −0.17 to −0.02], P = 0.010), but not in low-density lipoprotein cholesterol (−0.06 mmol/L [95% CI, −0.14 to 0.02], P = 0.170), high-density lipoprotein cholesterol (0.05 mmol/L [95% CI, −0.03 to 0.13], P = 0.220), or blood pressure (systolic blood pressure −0.82 mm Hg [95% CI, −4.65 to 3.00], P = 0.670; diastolic blood pressure −1.71 mm Hg [95% CI, −3.71 to 0.29], P = 0.090).
Conclusions
Among older adults with T2D, mHealth interventions were associated with improved cardiometabolic outcomes versus usual care.
Comments
This review examined the effectiveness of mHealth interventions on cardiometabolic outcomes in older adults with T2D. The main findings were (1) mHealth is an emerging forefront in patient-centric care; (2) the efficacy of mHealth interventions on health outcomes among older adults with T2D is limited; (3) the modest benefits on cardiometabolic outcomes with mHealth interventions have uncovered critical gaps in the field and offer insights to address barriers at hand. It is hoped that future studies will demonstrate impact. Patients and providers desperately need the help digital interventions can deliver.
Cost and Cost-effectiveness Analysis of a Digital Diabetes Prevention Program: Results from the PREDICTS Trial
Michaud TL1,2, Wilson KE3,4, Katula JA5, You W6, Estabrooks PA7
1Department of Health Promotion, College of Public Health, University of Nebraska Medical Center, Omaha, NE; College of Health, University of Utah, Salt Lake City, UT; 2Center for Reducing Health Disparities, College of Public Health, University of Nebraska Medical Center, Omaha, NE; College of Health, University of Utah, Salt Lake City, UT; 3Department of Kinesiology and Health, College of Education & Human Development, Georgia State University, Atlanta, GA; College of Health, University of Utah, Salt Lake City, UT; 4Center for the Study of Stress, Trauma, and Resilience, College of Education and Human Development, Georgia State University, Atlanta, GA; College of Health, University of Utah, Salt Lake City, UT; 5Department of Health and Exercise Science, Wake Forest University, Winston-Salem, NC; College of Health, University of Utah, Salt Lake City, UT; 6Department of Public Health Sciences, University of Virginia, Charlottesville, VA; College of Health, University of Utah, Salt Lake City, UT; 7Department of Health and Kinesiology, College of Health, University of Utah, Salt Lake City, UT
Technology-assisted diabetes prevention programs (DPPs) have been shown to improve glycemic control and weight loss, but information is limited regarding relevant costs and their cost-effectiveness. This retrospective within-trial cost and cost-effectiveness analysis (CEA) compared a digital-based DPP (d-DPP) with small group education (SGE) over 1 year.
Methods
The costs were summarized into direct medical costs, direct nonmedical costs (i.e., times that participants spent engaging with the interventions), and indirect costs (i.e., lost work productivity costs). The CEA was measured by the incremental cost-effectiveness ratio (ICER). The sensitivity analysis was performed using nonparametric bootstrap analysis.
Results
In the d-DPP group, the direct medical costs, direct nonmedical costs, and indirect costs per participant over 1 year were $4556, $1595, and $6942, respectively. In the SGE group, these costs were $4177, $1350, and $9204, respectively. Based on a societal perspective, the CEA results showed cost savings from d-DPP relative to SGE. Using a private payer perspective for d-DPP, ICERs were $4739 and $114 to obtain an additional unit reduction in HbA1c (%) and weight (kg), and were $19,955 for an additional unit gain of quality-adjusted life years (QALYs) compared with SGE, respectively. From a societal perspective, bootstrapping results indicated that d-DPP has a 39% and a 69% probability of being cost-effective at a willingness-to-pay of $50,000/QALY and $100,000/QALY, respectively.
Conclusions
The d-DPP was cost-effective and offers the prospect of high scalability and sustainability due to its program features and delivery modes, which can be easily translated to other settings.
Comments
Michaud and colleagues evaluated the costs associated with a digital Diabetes Prevention Program (d-DPP). They also examined the cost-effectiveness of the d-DPP with an enhanced usual care condition. The d-DPP was cost-effective in achieving HbA1c reduction and weight loss. The authors supported the prospect of high scalability and sustainability due to the program features and delivery modes, which can be easily translated to a variety of settings. These results, if replicated with other digital DPP interventions, justify the decision by the U.S. Centers for Medicare & Medicaid Services and others to pay for high-quality, research-proven digital DPP programs.
Summary
Several articles incorporated important concepts from implementation science, including one that offered a new web-based capacity planning tool to help diabetes centers match their ambition with digital health solutions to their resources. Several articles addressing participatory health or “connected care” platforms that support remote monitoring demonstrated the achievement of positive health outcomes using those platforms. If designed correctly, these platforms should increase the therapeutic alliance between patient and trusted health-care team.
The increasing number of randomized controlled trials is a positive sign that the quality of investigations in this field also is increasing. One article opted instead for a propensity score matching approach to define a control group for comparison in a real-world data study; this approach holds significant promise because it reduces the time and cost to arrive at initial insights regarding efficacy of a digital intervention. These concepts and themes indicate a vibrant field of inquiry, while the increasing volume of articles—still in a steep growth phase—indicates that there is likely increased support from funding agencies and increased commercial interest in digital health solutions.
So what is missing this year? We leave you with food for thought on the next frontier in this field: If glucagon-like peptide 1 (GLP-1)/glucose-dependent insulinotropic polypeptide (GIP) analogs continue to show dramatic success in the treatment of obesity and T2D, is it possible that digital companions—or at least digital health solutions promoting lifestyle interventions—can enhance the efficacy of these medications in a cost-effective way? After all, the recent SURMOUNT-3 study demonstrated that individuals who engaged successfully with a lifestyle intervention program (with > 5% weight loss over 12 weeks) achieved a 24.5% net weight change compared with placebo during a 72-week randomized trial of tirzepatide as well as a total weight loss of 26.6% from the beginning of the lifestyle intervention program (1). As new GLP-1–related therapies receive regulatory approval for the treatment of diabetes or obesity, the role of digital companions in maximizing drug benefit is likely to increase.
Footnotes
Author Disclosure Statement
Neal Kaufman is the founder of and has equity interest in Canary Health, LLC.
Mark Clements receives fees as Chief Medical Officer of Glooko, Inc. He has received research support from Dexcom and Abbott Diabetes Care.
Eran Mel has no competing financial interests.