An Evaluation of Digital Health Tools to Support Multidisciplinary Teams in Managing Patients with Type 2 Diabetes

Abstract

Type 2 diabetes requires sustained self-management and coordinated multidisciplinary care, with rising prevalence challenging health systems. Digital health tools and services have been introduced to support patients and healthcare professionals by extending access to education, behaviour change support, and data-enabled decision-making. Whilst previous reviews emphasise the effectiveness of digital interventions, this narrative review draws on systematic reviews, meta-analyses, trials, observational implementation studies, and qualitative evidence to synthesise what is known about their effectiveness, implementation in routine practice, and implications for multidisciplinary care. Randomised trials and observational studies evaluating web-based support, mobile applications, text messaging, and online peer support platforms suggest that digital tools may improve glycaemic control, medication adherence and health behaviours. However, effect sizes are variable, benefits are often short-term, and engagement typically declines over time. Evidence from real-world implementation is more limited and demonstrates a consistent “voltage drop” between trial efficacy and effectiveness in routine practice, driven by inconsistent delivery, low uptake, and limited integration into care pathways. Persistent inequalities in access and use of digital interventions linked to age, ethnicity, socioeconomic status, language, and digital literacy further constrain impact. Many interventions lack explicit theoretical foundations, rely on evidence generated from digitally literate populations, and offer limited cultural tailoring. Looking ahead, digital tools should be understood as adjuncts that can extend, reinforce, and personalise support between face-to-face contacts. Future research should prioritise real-world implementation, equitable access and engagement, and the effective integration of digital tools within multidisciplinary diabetes care. Realising this potential will require clearer referral pathways, shared understanding within multidisciplinary teams of which programmes are suitable for whom, and active support for patients to navigate, access, and engage with available options. Hybrid models that combine digital and face-to-face care may therefore be particularly important for embedding digital tools into routine practice while preserving person-centred clinical support.

Introduction

Type 2 diabetes mellitus (T2DM) remains one of the most significant global public health challenges, placing a substantial burden on individuals, health systems, and economies. More than 1 in 10 adults are living with diabetes,¹ and the global economic burden is projected to exceed 2.1 trillion dollars by 2030.² Effective management of T2DM includes improving nutrition, increasing physical activity and for many people with T2DM, increasing medication adherence.³ Without adequate support, people with T2DM face increased risk of serious complications, including retinopathy, neuropathy, limb amputation, and cardiovascular disease.³ Yet many of these outcomes are preventable when individuals are equipped to self-manage their condition effectively.

As prevalence of T2DM continues to rise, the complexity of care – encompassing regular monitoring of blood glucose levels and disease symptoms, as well as actions required to treat problems and prevent problems from worsening – requires coordinated input from multidisciplinary teams. In this review, multidisciplinary care refers to coordinated care delivered by healthcare professionals (HCPs) from different disciplines, including primary care clinicians, diabetes specialists, nurses, dietitians, psychologists, pharmacists, and other relevant professionals involved in supporting diabetes management. However, traditional models of care are increasingly strained, and gaps in continuity, communication, and patient engagement persist.

New opportunities to enhance the scalability, personalisation, and coordination of diabetes care have arisen through the increased availability of effective digital health technologies over the past decade. In this review, digital health tools refer to patient-facing and healthcare-supported technologies designed to support T2DM self-management, communication, monitoring, education, and decision-making. These include web-based programmes, mobile applications, text messaging interventions, remote monitoring technologies, and online peer-support platforms. Such tools can facilitate real-time data sharing, automated feedback, and more flexible interactions between patients and HCPs.⁴ These technologies hold promise for strengthening multidisciplinary team collaboration, supporting self-management, and empowering individuals to take a more active role in their care. Their value, however, depends on equitable access, sustained engagement, and integration into routine clinical workflows.

Numerous systematic reviews and meta-analyses have evaluated the effectiveness of specific digital interventions for T2DM, including web-based programmes (eg,⁵), mobile applications (eg,⁶), text messaging interventions (eg,⁷), and digital diabetes self-management education and support (eg,⁸). However, these reviews have typically focused on individual intervention modalities or clinical effectiveness outcomes, most commonly HbA1c and weight-related measures. Less attention has been paid to how digital tools for diabetes are implemented in routine practice, the factors influencing uptake and sustained engagement, implications for health inequalities, and how digital interventions can be integrated within multidisciplinary models of care. Consequently, there remains a need for a broader synthesis that brings together evidence to consider not only whether digital tools work, but how they can be successfully embedded into sustainable, equitable, multidisciplinary diabetes care.

This narrative review synthesises evidence from research studies and real-world application of digital tools and services for T2DM, examining their effectiveness, implementation challenges, and potential to support multidisciplinary care. It also identifies key gaps and outlines future directions to ensure that digital strategies meaningfully contribute to high-quality, coordinated diabetes management. As a narrative review, this article was designed to provide a broad, critical overview⁹ rather than an exhaustive systematic review or meta-analysis. Narrative reviews serve as an important, pragmatic function in knowledge translation and contextualising evidence within professional practice¹⁰ and are particularly appropriate where the aim is to summarise and interpret a heterogeneous body of literature, identify conceptual and practical issues, and generate implications for research and practice.¹¹

In line with published typologies of review methods,^10,11 we used a structured but flexible literature identification approach to locate relevant evidence on digital tools for T2DM self-management and their implications for multidisciplinary care. We prioritised recent systematic reviews and meta-analyses where available, supplemented by key randomised trials, observational and implementation studies, qualitative research, and relevant policy or service evaluation evidence. Literature was identified through targeted database searches, citation tracking, author expertise, and searches for recent evidence on specific intervention modalities. Because this was not a systematic review, we did not conduct formal quality appraisal or meta-analysis;¹¹ instead, we narratively synthesised evidence according to intervention modality, implementation context, equity considerations, and relevance to multidisciplinary care.

This narrative review therefore aims to synthesise current evidence on digital tools for T2DM self-management and consider their implications for implementation, service delivery, and multidisciplinary care. Examples from diabetes prevention and wider digital health literature are occasionally discussed where direct evidence is limited; these are used to inform understanding of implementation, engagement, and service delivery considerations rather than as evidence of effectiveness in T2DM management itself. Unlike previous reviews that have predominantly focused on effectiveness of specific digital health tools, this review integrates evidence relating to effectiveness, implementation, engagement, equity, and service delivery. In doing so, it seeks to identify key challenges, evidence gaps, and priorities for future research and practice to support the effective integration of digital tools within routine multidisciplinary diabetes care for T2DM.

Self-Management of Type 2 Diabetes

Self-management of T2DM is multifaceted and involves people with T2DM carrying out a number of actions to detect worsening of blood glucose levels and symptoms, seek appropriate help when these occur, and take actions to prevent deterioration. Although these behaviours are essential for maintaining health, many people perceive them as burdensome or disruptive to quality of life and therefore do not consistently follow medical recommendations.¹²

A core component of self-management involves adhering to recommended monitoring. Some monitoring occurs during routine medical appointments, such as diabetic eye screening for signs of diabetic retinopathy, foot examinations for neuropathy and peripheral vascular disease, and checks of cardiovascular risk factors, given the elevated cardiovascular risk associated with T2DM. Individuals who develop complications or concerning symptoms may also be referred to secondary care services, requiring additional attendance and follow-up.

Checking of long-term blood glucose levels (HbA1c) will happen at medical appointments, but people with T2DM are also encouraged to monitor short-term variations in blood glucose between visits. Such monitoring can be done using finger-prick tests but given that these are not acceptable to many people,¹³ instead continuous glucose monitoring devices that sit under the skin and provide a more automated assessment are increasingly being used in some groups of people with T2DM, particularly those treated with insulin or at greater risk of hypoglycaemia,¹⁴ although access varies across healthcare systems and patient populations. Self-monitoring can increase the awareness of people with T2DM regarding what factors such as physical activity and diet affect glucose levels, as well as providing information on glucose trends and low-glucose episodes, which may be particularly beneficial for individuals at increased risk of hypoglycaemia (where blood glucose levels drop to an abnormally low level). In this review, we focus on continuous glucose monitors as peripherals, providing data for digital diabetes tools rather than as interventions in and of themselves.

Preventive behaviours form another major element of T2DM self-management. Many of these behaviours are in line with those recommended to the general public, such as eating a diet high in fibre, and being physically active as well as not smoking or limiting the amount of alcohol consumed. These behaviours are particularly important to people with T2DM, given their higher chances of cardiovascular disease and peripheral vascular disease. Specific dietary guidance often emphasises a Mediterranean diet, including oily fish and low in saturated fats. Managing carbohydrate consumption is also crucial, as foods such as bread, potatoes and biscuits can significantly affect blood glucose. People with T2DM must maintain a balance between intake of carbohydrates and activity levels, to avoid elevated blood glucose levels over long periods of time on the one hand, but hypoglycaemia on the other hand. Adhering to these recommendations is a challenge for many people, whether they have T2DM or not, but there appear to be particular problems for people with T2DM, due to confusion over exactly what they should or should not be eating.¹⁵

Medication adherence is another important aspect of self-management. Many people with T2DM who are prescribed medication take less than prescribed or do not take it at all.¹⁶ There are a number of factors that influence medication taking, ranging from forgetting to more deliberate choices to alter or discontinue medication.^17,18

An often-neglected category of self-management concerns the psychological impact of being diagnosed with, and living with, T2DM.¹⁹ Approximately 36% of people with T2DM experience significant psychological distress.²⁰ Apart from being of importance in its own right, issues such as diabetes-related depression have been shown to be associated with poorer self-care, higher blood glucose levels, increased risk of severe hypoglycaemia and reduced quality of life.^19,21,22 Beyond emotional management, individuals must also navigate changes in life roles, relationships with significant others, working patterns, and their capacity to carry out routine activities.²³ Clinician visits can be infrequent for those with T2DM; it is therefore imperative to effectively support people to make and maintain self-management behaviours between visits.

Digital Diabetes Tools: Evidence from Trials and Observational Studies

Evidence from systematic reviews and meta-analyses suggests that digital interventions can produce modest improvements in glycaemic control, dietary behaviours, and weight-related outcomes among people with T2DM (eg,^5,24–27). However, effect sizes vary substantially across intervention types, study populations, and levels of engagement and interventions are heterogeneous including variation in mode of delivery, content and types of data used, making synthesis challenging. The majority of this evidence relates to patient-facing self-management interventions and clinical or behavioural outcomes. Direct evidence regarding multidisciplinary team functioning, workflow, workforce implications, and service delivery remains comparatively limited. This section therefore focuses on the evidence for specific types of digital tools and services, while later sections address cross-cutting issues relating to theory, equity, implementation, and multidisciplinary care.

Web-Based Support

Evidence for web-based interventions derives primarily from randomised controlled trials (RCTs) and systematic reviews, with fewer studies evaluating long-term effectiveness in routine clinical practice. Web-based support can deliver a large volume of evidence-based content that can be accessed at any time and tailored to users’ interests or available data. Users can receive personalised support using their own data from peripheral devices such as continuous glucose monitors, activity trackers and weighing scales. In addition, routine data such as appointments, medication schedules and test results can also be incorporated. Web-based interventions can present information as well as incorporating interactive elements such as tools to support behaviour change and support communication between people with T2DM and their HCPs and peers. Systematic review evidence suggests web-based interventions can be effective in supporting self-monitoring of HbA1c²⁸ with RCTs and real‑world evaluations suggesting they can reduce HbA1c^29,30 with stronger effects among users who engage more frequently,^30,31 use the advanced features such as direct messaging³² or access a greater amount of the available content.³³

Web-based support is increasingly available through a range of devices such as computer, laptops, tablets and smartphones. Not requiring a computer can increase the reach.²⁶ Systematic review evidence suggests that mobile applications can support improvements in HbA1c among people with T2DM,⁶ and they may result in greater reductions than computer-based approaches.³⁴ Although these findings are based on small samples and in real-world implementation settings the results are more mixed due to low use and engagement.³⁵

In the United Kingdom (UK) in 2024, 93% of people aged 55–64 and 77% of people aged 65 and over owned a smartphone,³⁶ and in the United States of America, 91% of adults were reported to own a smartphone in 2025.³⁷ This is in stark contrast to low‑ and middle-income countries, where overall mobile phone ownership is high, but access to internet-capable smartphones is 52%.³⁸ In addition, not everyone who owns a smartphone will feel confident or be able to download and use an app. Furthermore, reliance on mobile data or signal can disadvantage those living in rural areas.³⁹ Similarly to websites, there is the risk of exacerbating inequalities as individuals with certain characteristics for example, people who are younger, White and have higher levels of education are more likely to use apps and the more advanced features of apps such as linking to electronic health records.³¹ Evidence from qualitative studies suggests that HCP recommendation can encourage app use, but many HCPs do not routinely recommend apps or feel sufficiently knowledgeable to support their use.³⁹ Further research is needed to explore how the potential elements of web-based support act together to produce results and how to optimise their use within multidisciplinary care teams. These issues are revisited in later sections on implementation and equity.

Diabetes Self-Management Education and Support

In the UK, online Diabetes Self-Management Education and Support (DSMES) courses commissioned by the National Health Service (NHS) follow a structured format in which users progress through content in a pre-specified order, with later content released when earlier content has been viewed/completed. This has clear advantages in terms of fidelity of delivery of evidence-based content;³³ however, evaluations of such programmes in routine care suggest that users do not access content that might be key to changing behaviour⁴⁰ or that might be more relevant for them if it is presented later in the course.⁴¹ A particular example is the DSMES “Healthy Living” where support for emotional management appears later in the schedule. When interviewed after using Healthy Living, people who used the emotional management section really valued it, and for some this prompted a discussion about low mood with their HCP, however, others who said they would have been interested and found it beneficial did not know this content was available as they did not continue the course long enough to see it.⁴¹ This shows the potential of digital tools to support the multidisciplinary team to provide relevant care, but also that this potential can be missed when content is not accessed.

The alternative, to allow users to access the material that is of most interest to them, can similarly result in key mechanisms for behaviour change not being delivered (eg,^40,42) and therefore the course not being as effective as it could be. Further research is needed to explore the optimal balance between fidelity to evidence-based content and tailoring for web-based interventions, how to encourage optimal use across the population of people with T2DM and how digital interventions can be used to support multidisciplinary teams to support individuals.

Incorporating Continuous Glucose Monitors Within Web-Based Support

There is emerging RCT evidence that incorporating continuous glucose monitors within diabetes education offered face-to-face, or face-to-face with additional phone support, could be effective for reducing HbA1c and improving dietary and physical activity behaviours for people with T2DM.^43–45 However, integration into wholly web-based support has not been widely studied. One RCT has shown positive effects of combined digital education, one-to-one remote support from a coach and continuous glucose monitoring on HbA1c compared to usual care but reported some problems with adherence to the continuous glucose monitoring and people accessing the educational content.⁴⁶

Text Messaging

Text messaging interventions have low cost per person and high reach requiring neither the internet nor the expertise required to download and use an app.⁴⁷ They can provide behavioural support as well as simple reminders for appointments and medication.⁴⁸ Being low burden for the user means lower engagement is needed from the user to benefit. This is an advantage,⁴⁹ but the content by necessity must be brief.

Qualitative evidence suggests using text messages is acceptable to a diverse range of people with T2DM^50,51 and 78% of the worldwide population have a phone capable of receiving text messages.⁵² Systematic review and meta-analytic evidence suggests that text messaging interventions may improve HbA1c and support dietary, physical activity, and medication-taking behaviours, although effect estimates vary considerably across studies and intervention designs.^7,53,54 Recipients of text messages report feeling cared for and kept on track by the consistent low-level nudges.⁴⁹ Although potential of such an approach has been identified, interventions often have short follow-up, small samples, and do not describe the theoretical basis of the text messages used.⁷ Transparent reporting of the proposed mechanism of action for text message interventions would help to explore which components might be associated with effectiveness.

Secure messaging systems within patient portals and electronic health records are receiving increasing attention and greater use of these systems is associated with improvements in HbA1c. However, overall use remains low, with much of the evidence correlational.^55–57 Bi-directional message systems can increase contact with HCPs, which is something people with T2DM have said they would value.⁴⁹ However, HCPs report that while messaging is good for routine tasks, some messages do not contain sufficient information to respond to and can result in time consuming back and forth with the patient.⁵⁸ Integrating SMS messages into wider health systems might be an advantage as factors associated with greater effectiveness are combining text messages with other modes of intervention delivery^59,60 and using them to facilitate communication with the healthcare team,²⁵ however, further evidence is needed to explore characteristics of those more likely to use secure messaging systems and which elements within the messages and/or the portals are responsible for the positive effects, to inform how this technology can be best used to support the multidisciplinary care team.

Online Discussion Forums and Peer Support

Online discussion forums can provide peer support, companionship, information and self-management support⁶¹ and can be offered within larger online education programmes, or are available through social networking sites such as Facebook or Reddit and through social media apps such as WhatsApp or WeChat. They differ in the extent that they are moderated by either HCPs or by peer facilitators and some provide structured prompts to encourage engagement while others are flexible to allow conversation on any topic. Evidence relating to online peer support is derived largely from observational, cross-sectional, and qualitative studies, meaning that causal effects on clinical outcomes remain uncertain.

Cross-sectional studies have found correlations with engagement with this type of support and higher self-efficacy,⁶² and positive effects on empowerment and quality of life.⁶³ The effects on health outcomes are mixed, and the design of the studies thus far has meant the direction of influence is not clear, with the potential that those with higher health needs may be more likely to seek this type of support. In addition, many studies have short follow-ups and use self-selected samples.⁶²

Online groups supported by charities such as Diabetes UK in the UK are often moderated, so that any inaccurate, misleading or inappropriate content can be removed. Although concerns have been expressed about the accuracy of information provided by online groups⁶⁴ in one study, 91% of the recommendations given reflected clinical guidelines⁶⁵ and fora have been found to “self-police” inaccurate information meaning the risk of negative consequences of using online support is thought to be low.⁶³ Evidence from the diabetes prevention and self-management literature suggests that personal preference for this type of communication, how active a forum is, and HCP support are likely to influence engagement.^66,67 Further research could explore the potentially distinct role in supporting behaviour change online fora and peer support have when offered independently or as part of a wider programme,⁶⁸ clarify how online support is best facilitated, who might be most likely to benefit from it, and what optimal engagement looks like so this can be supported.⁶⁹

New Approaches to Address Emotional and Psychological Needs

A growing body of work is exploring “new” psychological approaches for T2DM to address the emotional and psychological effects of living with T2DM. Digital acceptance and commitment therapy (ACT) currently has the strongest evidence among these newer approaches. However, current evidence is derived primarily from systematic reviews and relatively small, randomised feasibility trials, with limited evidence regarding long-term effectiveness or implementation in routine care. ACT aims to increase psychological flexibility by fostering acceptance of unchangeable aspects of life, such as living with diabetes, rather than directly targeting behaviour change. Systematic reviews and meta-analyses of ACT for T2DM (not limited to digital delivery, but including some technology-supported formats) report significant improvements in HbA1c, self-care behaviours, diabetes acceptance, self-efficacy, and psychological outcomes such as anxiety and depression, although the certainty of evidence varies across outcomes.^70,71 Internet- and mobile-based ACT programmes have been developed specifically for people with diabetes which have shown feasibility and preliminary efficacy in reducing diabetes distress.^72,73

Mindfulness-based digital interventions form a smaller and more heterogeneous evidence base. Systematic review evidence suggests that therapist-led mindfulness- and acceptance-based approaches may reduce distress, HbA1c levels and promote self-care in people with T2DM.⁷⁴ Mindfulness is therefore a plausible approach for targeting stress, emotional eating, and sleep, but its effectiveness in digital T2DM self-management remains under-explored.

Together, ACT and mindfulness-based approaches may help address the psychological and emotional dimensions of living with T2DM that are often insufficiently targeted by conventional digital tools. However, more robust, large-scale RCTs are needed to evaluate effectiveness and explore how these approaches can be integrated into multidisciplinary care without increasing burden for patients or HCPs.

Just-in-Time Adaptive Interventions

A new technology that shows promise is just-in-time adaptive interventions (JITAIs), which can deliver tailored behaviour change support to people at moments when individuals are most at risk of engaging in behaviours misaligned with their health goals.⁷⁵ Tailoring and delivery can be triggered automatically based on real-time data, without requiring user input. JITAIs therefore offer potential for more timely, personalised, and context-sensitive support. Although no studies have yet focused specifically on JITAIs for T2DM self-management, early work in weight management⁷⁶ and physical activity promotion⁷⁷ is promising, and their potential in T2DM self-management should be explored.

Digital Diabetes Tools in Routine Practice

What is Currently Implemented?

Whereas the preceding section focuses primarily on evidence relating to patient-facing interventions and self-management outcomes, this section considers the implementation of digital tools in routine practice and their implications for HCPs, services, and multidisciplinary care pathways. The UK provides a useful exemplar for illustrating referral and navigation challenges because it has a relatively centralised healthcare system, national policy support for structured diabetes education, and a wide range of face-to-face and digital diabetes services. If challenges relating to referral, navigation, engagement, and sustained implementation remain evident in this context, they are likely to be at least as important in less coordinated or more fragmented healthcare systems.

Evidence from systematic reviews indicates that DSMES programmes are associated with improvements in both clinical outcomes and psychosocial wellbeing, as well as reductions in healthcare expenditure.^8,78 For instance, one of the most widely used programmes in England is “DESMOND”, a structured group education programme for people with newly diagnosed T2DM. DESMOND has demonstrated effectiveness at outcomes including weight loss in trials,⁷⁹ but also at reducing long-term blood glucose levels for patients who took up the programme as part of routine practice.⁸⁰ Despite this evidence, participation in traditional face-to-face DSMES remains low worldwide.⁸¹ Digital modes of delivering DSMES may help to overcome some of the practical barriers associated with attending in-person programmes, such as scheduling constraints, travel requirements, and competing work or caregiving responsibilities.^82,83

How are People Referred to These Programmes?

Low uptake is partly explained by referral pathways. People with T2DM are typically referred to digital tools and services by HCPs, but qualitative research has highlighted that HCPs forget to offer them to patients due to time pressures and report a lack of awareness about their benefits to patients.⁸⁴ Importantly, HCPs are not clear which T2DM service to offer patients, and confusion in understanding which types of patients were suitable for which programme.⁸⁵ This confusion is understandable, given the large number of available self-management services and the variation in provision across localities. For instance, in England there are structured group education in-person sessions in local communities (eg, “X-PERT”, “DESMOND”), as well as digital programmes (eg, “MyWay Diabetes”), and national programmes such as the NHS “Healthy Living” programme, the NHS “Type 2 Diabetes Path to Remission” programme and the “NHS Digital Weight Management Programme”. There are also numerous culturally tailored programmes available in different geographical regions. This means that supporting patients to use digital tools cannot be reduced to simply signposting them to a website or app. HCPs need clear referral criteria, up-to-date knowledge of locally and nationally available programmes, and confidence to explain why a particular option may be relevant to the patient’s needs, preferences, language, and self-management goals. Systematic reviews highlight that successful implementation of digital interventions in primary and secondary care requires organisational readiness, including clear clinical roles, adequate resources, integration with existing workflows, and ongoing monitoring to ensure interventions achieve their intended outcomes.^86–89

At the point of referral, HCPs can play an important role in legitimising digital tools and services as part of routine diabetes care. This may include introducing the programme during consultations, helping patients understand what the programme is intended to support, clarifying how it complements rather than replaces face-to-face care, and checking whether the patient anticipates practical barriers such as access to devices, data, confidence, language, or competing demands. Where possible, referral pathways should include brief onboarding support, such as demonstrating how to register and use the programme, and agreeing how digital tool use will be reviewed at future appointments.⁹⁰ Such support may be particularly important for patients who are less confident with technology or who have previously disengaged from structured education.

What is the Evidence of Their Effectiveness in Routine Practice?

It is important that programmes are able to demonstrate effectiveness in routine practice as well as in trials, given that there is often “voltage drop” in effectiveness when such programmes are delivered in routine practice.^91,92 Understanding the reasons for this drop in effectiveness is essential for multidisciplinary teams responsible for implementation and support and to continue to optimise digital health technologies once implemented.^93,94

A major issue of “scaling up” interventions from research studies to real-world interventions is that it can entail loss of fidelity to the original intervention, in terms of specific contents that are informed by its underlying theoretical basis. A worked example from the NHS Diabetes Prevention Programme in England has shown how there was a loss of fidelity at multiple stages, from the design of the intervention deviating from the underlying evidence base, through training of providers, to delivery by providers, and onto understanding of the intervention and carrying out the intervention activities by participants.⁹⁵ There are multiple reasons for this loss in fidelity, but having a clear logic model that explicitly sets out how an intervention should work and why, and contextual influences on this can be useful in helping maintain fidelity, for example, in training of providers.⁹⁶ Although derived from the diabetes prevention literature rather than T2DM self-management, these findings highlight potential challenges relating to fidelity during scale-up. Reporting fidelity of the intervention delivery further helps establish explanations for differences in effectiveness between trials and real-world implementation and whether changes to intervention design or the process of delivery might be needed.⁹⁷

The flip side of fidelity is adaptation, and it may often be the case that interventions are less suited for people with T2DM in real-world settings than they are for people in trials. It is a reliable observation that people in trials tend to differ from people for whom interventions are designed, and this can impact on effectiveness. For example, there is increasing evidence that behavioural interventions work less well with older people than the younger people they are typically developed with, at least partly due to cognitive abilities declining with age.⁹⁸ Other such moderating factors include people in trials being healthier with less co-morbidity, more motivated, more educated and higher socioeconomic status, and less likely to belong to an ethnic minority group. Thus, the interventions can feel less suitable for the intended target population. For these reasons, engagement in interventions outside of a trial context can be lower,⁹⁴ although this can still yield useful effects for those people.³⁰

In routine practice, HCP support is also needed after referral. Digital programmes often require patients to interpret information, set goals, monitor progress, and translate generic advice into everyday decisions about food, activity, medication, and help-seeking. HCPs can reinforce the intended behaviour change content by asking about patients’ use of digital tools during reviews, helping them interpret feedback or patient-generated data, troubleshooting barriers to engagement, and linking digital content to personalised clinical priorities. Multidisciplinary teams therefore need agreed processes for who reviews engagement or alerts, how patient-generated data are interpreted, and when digital concerns should trigger clinical follow-up.

The increasing availability of patient-generated data from apps, wearable devices, continuous glucose monitors, and AI-enabled systems also raises important questions regarding governance, clinical responsibility, and workload. Digital tools may generate large volumes of information, but multidisciplinary teams require clear processes to determine which data are reviewed, by whom, how frequently, and what actions should follow. Without clear role allocation, there is a risk of duplication of work, inconsistent responses, unclear accountability, and increased burden on HCPs.⁹⁹ Future implementation research should therefore examine how digital tools can be integrated into existing workflows in ways that support clinical decision-making while remaining feasible, safe, and sustainable for multidisciplinary teams.

How Could Digital Diabetes Tools Be Improved?

Although digital interventions for T2DM show considerable promise, their effectiveness is constrained by several limitations in how they are designed, reported and evaluated, including the integration of evidence-based content, explicit use of theory to guide intervention development, and the transparent description of included components and proposed mechanisms of action.

Better Use of Explicit Theory and Evidence-Based Content

A consistent limitation across digital interventions for T2DM is the lack of explicit theory guiding their design while theory-based techniques are associated with greater intervention effects.²⁴ Evidence consistently highlights the importance of particular behaviour change techniques (BCTs) to support self-regulation of behaviours including goal setting, self-monitoring, feedback, and social support to be effective in changing dietary and physical activity behaviours to reduce HbA1c and weight in people with T2DM,¹⁰⁰ and in technology-driven interventions for diabetes prevention.¹⁰¹ Similarly, a recent meta-analysis of digital behaviour change interventions for adults with T2DM reported that interventions using more than ten BCTs, delivered over longer durations and often combining digital and face-to-face elements, produced significantly greater improvements in physical activity.¹⁰² Likewise, a systematic review of digital coaching interventions for T2DM also found that effective programmes typically integrated personalised feedback, education, goal setting, and ongoing behavioural support, although it did not formally code BCTs.¹⁰³ Despite this progress, less is known about how two or more BCTs interact, the timing and “dose” required, or whether specific techniques are more influential for particular populations or contexts. Addressing these gaps will require detailed reporting of intervention content and more sophisticated analyses of mechanisms of action.

A further challenge concerns the representativeness of the evidence base. Much of what is known about effective intervention components comes from studies involving younger, more educated, ethnically homogeneous, and digitally literate participants.¹⁰⁴ These groups may find self-regulatory content such as goal setting, self-monitoring, and planning easier to engage with than individuals with lower health literacy, limited digital skills, or more complex social circumstances. For example, qualitative research with service users of a nationally implemented digital diabetes prevention programme in the UK revealed that understanding of some key BCT content (eg, action planning, problem solving) was often poor, with many service users having little recollection of these BCTs.¹⁰⁵ More inclusive research is therefore needed to ensure that evidence-based content is genuinely applicable across the diverse groups living with T2DM. This includes understanding how different populations interpret, engage with, and benefit from specific BCTs, and ensuring that digital interventions are designed with the needs of those most at risk of being left behind.

Clear Reporting of Intervention Content

Although digital tools can improve glycaemic outcomes, many studies do not clearly explain how behaviour change and self-management are expected to occur and what proposed mechanisms are included.⁷ Under-reporting of intervention content limits interpretation, replication, and identification of the components responsible for observed effects.¹⁰⁶

Greater use of logic models or explicit descriptions of how interventions are expected to work¹⁰⁷ would help establish the key techniques and mechanisms of action underpinning digital interventions for T2DM.⁹⁶ Standardised frameworks to describe intervention content can further support this. The Template for Intervention Description and Replication (TIDieR) checklist¹⁰⁶ provides a structured approach to reporting intervention components, while tools such as the Behaviour Change Technique Taxonomy¹⁰⁸ and the more recent Behaviour Change Technique Ontology¹⁰⁹ enable precise description of the active behaviour change components. Without clear reporting of intervention rationale and behaviour change content, systematic reviews are severely constrained in testing which BCTs or other features work in changing health behaviours and why.

Greater use of explicit theory, evidence-based content and more transparent reporting would improve intervention coherence by ensuring that digital features: are purposefully selected to target specific behavioural determinants; enhance fidelity and replicability by enabling multidisciplinary teams to understand how and why an intervention works; support more rigorous evaluation by allowing researchers to test mechanisms of action; and facilitate adaptation and scaling by enabling developers to modify components while preserving core mechanisms.¹¹⁰ For multidisciplinary teams, more transparent reporting and greater understanding of mechanisms of action could lead to more targeted, personally relevant referrals. In addition, this would enable HCPs to identify and reinforce key components of the digital tools and services during follow-up appointments, supporting longer-term engagement with the tools creating a feedback loop between the tools and HCP.

Key Challenges and What Can Be Done About Them

Long Duration Between Development and Implementation

We have outlined the importance of using research evidence to integrate behaviour change theories and techniques into digital health interventions. However, this evidence-based development takes time and resources, and there is a mismatch between the speed at which new, potentially useful, technologies are developed and the design, evaluation and implementation of interventions that use these technologies.

One potential way to reduce the time needed to develop interventions is to adapt existing interventions to new contexts.¹⁰⁷ Approached systematically, with understanding of the initial and new contexts, adaptation can enable researchers to increase efficiency while maintaining how the intervention functions in a new context. For this to be successful, careful exploration of key areas of difference to ensure acceptability to a new target population is needed.¹¹¹ In addition, considering staff workflow in new contexts is needed to ensure introduction of the technology is appropriately supported, integrated into existing systems, and does not inadvertently increase staff workloads.¹¹² Examples of this approach include adapting a nationally implemented web-based self-management programme from the UK to Australia¹¹³ and adapting a diabetes prevention programme from the US to South Africa.¹¹⁴

Potential to Increase Inequalities

Digital technologies can increase access for people with T2DM facing barriers to in-person care, including those living in rural locations, those with mobility problems and those with caring responsibilities. However, differences in access, confidence, and ability to use digital technologies persist across high income countries¹¹⁵ and are further pronounced in low- and middle-income countries (eg,¹¹⁶).

Data related to diabetes from wearable technologies, continuous glucose monitors and patient health records are increasingly available and new tools and models of diabetes management that utilise artificial intelligence (AI) and machine learning have the potential to improve prediction of both the development of diabetes, its complications, as well as personalising treatment plans.^117–119 However, there is a risk that as systems become more complex, they become less easy to use,¹¹⁸ which could exacerbate existing inequalities and make it harder for HCPs to offer support with their use.¹¹² In addition, missing data from routine records are not equally distributed across populations with people from more deprived backgrounds and with larger disease burden for example more likely to have incomplete data.¹²⁰ In developing models based on routine data, there is a risk these models will not reflect the diversity of the people with diabetes. Exploring acceptability and appropriateness of the interventions to target populations as well as usability to both HCPs and people with diabetes remains central to future developments.^118,121

Supporting Equity in Structured Diabetes Education Programmes

Some populations such as older adults, those from minoritised ethnic groups or those with learning disabilities having lower uptake of digital health interventions.⁸³ Furthermore, disadvantaged groups have been shown to engage with digital DSMES less and abandon them more quickly.^94,122 Real-world studies of nationally implemented T2DM self-management programmes have shown that lower engagement can result in reduced exposure to key behaviour change content.^40,94 These differences in use could lead to less opportunity to effectively support behaviour change in certain groups and differences in outcomes achieved. The challenge remains therefore to utilise technology in a way that does not increase existing inequalities in care and outcomes.

Several strategies have been proposed to support uptake, engagement, and sustained use across diverse populations. These include integration within existing health services, supportive and knowledgeable HCPs and provision of tailored content with acceptability to traditionally underserved groups.^{83,118,123,124} In addition, specific features designed to increase ongoing engagement with T2DM self-management programmes such as goal setting and provision of feedback⁹³ could be made available from the outset, rather than being part of a structured process meaning that those who disengage early would still receive them.⁴⁰ Offering additional content to support engagement alongside the programme using multiple modes of delivery such as text messages could also be explored.⁵⁰ The content of messages could be co-created with specific populations, addressing the specific barriers faced to potentially increase acceptability and longer-term engagement with the programmes.

For HCPs, supporting equity means proactively identifying who may need additional help to access and use digital tools, rather than assuming that non-use reflects lack of motivation. This could include checking digital access and confidence as part of routine diabetes reviews, offering assisted registration or demonstration, signposting to translated or culturally tailored content, involving family members or carers where appropriate, and ensuring that non-digital or blended alternatives remain available. HCPs and service providers can also monitor whether particular groups are less likely to be referred, register, or continue using digital programmes, enabling teams to adapt referral processes and support offers before inequalities widen.

Conclusion and Future Directions

Digital tools can support T2DM self-management and have the potential to enhance multidisciplinary care through education, self-monitoring, tailored feedback, reminders, peer support, and communication with HCPs. However, while evidence for patient-facing self-management interventions is growing, direct evidence relating to multidisciplinary team outcomes, workflow, workload, and service delivery remains comparatively limited. Table 1 summarises the evidence across digital tools. The evidence suggests potential benefits for glycaemic control, medication adherence, health behaviours, self-efficacy, and emotional management of diabetes, but effects remain variable and are shaped by intervention content, delivery, engagement, and context. This review highlights a gap between trial efficacy and routine implementation. Digital interventions require clearer theoretical foundations, transparent reporting, fidelity assessment, and pathways that support sustained, equitable use in practice. Their impact is likely to be greatest when embedded within multidisciplinary care, supported by knowledgeable HCPs, and responsive to patient preferences and local contexts.

Table 1.

Summary Comparison of Digital Interventions for Type 2 Diabetes Management

Intervention Category	Evidence Source	Typical Characteristics	Key Strengths	Key Limitations	Summary of Evidence
Web-based platforms	Systematic reviews, RCTs and observational real-world studies	Online education, DSMES programmes, portals, self-management resources, interactive modules, messaging, and access to personal health data. Many are available through mobile phones.	Can deliver structured evidence-based content at scale and enable tailoring, self-monitoring, communication, and access outside clinic visits. Delivery via mobile phones makes them portable Continuous glucose monitors can be incorporated for self-monitoring.	Requires internet access, confidence, and sustained engagement; structured sequencing may limit exposure to key content if users disengage early.	Moderate to strong evidence from systematic reviews and trials for glycaemic and behavioural outcomes, with effectiveness dependent on engagement and implementation context. Growing evidence from RCTs for peripherals, eg, continuous glucose monitors.
Text messaging and secure messaging interventions	Systematic reviews, meta-analyses, RCTs and qualitative studies	Automated or tailored text messages, reminders, motivational prompts, medication support, appointment reminders, and bidirectional patient-HCP messaging.	Low cost, high reach, low user burden, and does not require smartphone apps or broadband access; can provide regular prompts and perceived support.	Content must be brief; effects vary by tailoring, intensity, and integration with wider care; bidirectional messaging can increase HCP workload.	Moderate evidence for improving medication adherence, lifestyle behaviours, and HbA1c, although studies are often short and intervention mechanisms under-reported.
Online peer support and discussion forums	Observational and qualitative studies	Moderated or unmoderated online communities, social media groups, discussion boards, or group chats offering peer advice, emotional support, and shared experience.	Can reduce isolation, increase empowerment, support experiential learning, and complement professional advice.	Quality and accuracy of advice may vary; samples are often self-selected; benefits may depend on moderation, group activity, and personal preference.	Emerging to moderate evidence for psychosocial benefits, self-efficacy, and empowerment; weaker evidence for direct clinical outcomes.
Acceptance and Commitment Therapy and mindfulness	Systematic reviews and small RCTs	Digital or technology-supported psychological interventions designed to increase psychological flexibility, acceptance, mindfulness, emotional regulation, self-care, and coping with diabetes distress.	Addresses emotional and psychological needs that conventional self-management tools may not sufficiently target; may improve distress, acceptance, self-efficacy, self-care, and psychological outcomes.	Digital T2DM-specific evidence remains limited; studies are often small or feasibility-focused, with limited long-term effectiveness and implementation evidence.	Emerging evidence suggests ACT and mindfulness-based approaches may improve HbA1c, self-care, acceptance, self-efficacy, distress, anxiety, and depression, but larger trials and implementation studies are needed.
AI-assisted tools and predictive analytics	Scoping reviews, early evaluations, routine-data modelling, implementation literature, and emerging feasibility evidence.	Machine learning models, chatbots, conversational agents, clinical decision support, risk prediction tools, and personalisation algorithms using patient-generated or routine data.	Potential to personalise support, predict risk, optimise treatment, triage need, and integrate complex data from records, wearables, and sensors.	Evidence is still developing; concerns include bias, transparency, safety, privacy, incomplete data, usability, accountability, and maintaining human oversight.	Emerging evidence, mainly feasibility, scoping, observational, and early evaluation studies; robust trials and real-world implementation evidence are needed.

Abbreviations: ACT, Acceptance and Commitment Therapy; AI, artificial intelligence; DSMES, Diabetes Self-Management Education and Support; HCP, healthcare professional; RCT, randomised controlled trial; T2DM, type 2 diabetes mellitus.

This review has identified several digital approaches that show potential to support T2DM self-management; however, evidence regarding the magnitude and real-world sustainability of effects remains variable across intervention types and settings. Further, these benefits will not be realised if people with T2DM are not referred to appropriate tools or are not supported to use them in ways that are meaningful for their everyday self-management. This represents a key role for HCPs and multidisciplinary teams, particularly in the context of rising T2DM prevalence and increasing pressure on healthcare services. By actively identifying suitable tools and services, making timely referrals, supporting onboarding, and revisiting digital tool use during routine care, HCPs can help ensure that digital technologies extend rather than fragment care, increasing the reach of evidence-based support while preserving the relational and personalised aspects of diabetes management.

This review has limitations. As a narrative review, it did not use systematic review methods, formal risk-of-bias assessment, or meta-analysis. The literature identification approach was structured but not exhaustive, and the synthesis necessarily reflects interpretive judgements about relevance across a heterogeneous evidence base. Some implementation and behaviour change insights discussed in this review derive from diabetes prevention or wider digital health literature where direct evidence relating to T2DM was limited; therefore, transferability to all T2DM care settings should be interpreted cautiously. In addition, much of the evidence concerns patient-facing self-management interventions, with more limited direct evidence on multidisciplinary team outcomes, workflow, workload, safety, accountability, and long-term implementation. Nonetheless, this review extends existing evidence syntheses by moving beyond intervention effectiveness alone to consider the practical challenges of implementation, sustained engagement, health equity, and integration within multidisciplinary models of care, thereby providing a broader perspective on how digital tools can contribute to routine T2DM management.

Table 2 summarises future work for research and multidisciplinary care. Inclusive co-design, accessible content, flexible support models, and careful evaluation of emerging technologies such as AI-enabled decision support, continuous glucose monitoring, and wearables will be essential if digital tools are to contribute to more accessible, personalised, and coordinated T2DM care without widening inequalities. Future research should prioritise long-term real-world effectiveness, implementation within routine multidisciplinary care, and the equitable integration of emerging digital technologies to ensure that benefits are sustainable, scalable, and accessible to diverse populations.

Table 2.

Future Priorities for Research and Multidisciplinary Care Teams Using Digital Health Tools for Type 2 Diabetes Management

Priority Area	Future Research Priorities	Implications for the Multidisciplinary Care Team
Which digital tools and services are suitable for whom	Consensus process informed by reviews of the evidence to agree which digital tools and services are most useful for which groups of people.	Receive up-to-date knowledge of locally and nationally available programmes to explain why a particular digital tool or service may be relevant to the patient’s needs, preferences, language, and self-management goals.
Referral, onboarding, and supported use	Examine how referral processes, onboarding support, and routine follow-up influence uptake, meaningful engagement, and sustained self-management benefits in real-world T2DM care without increasing administrative burden.	Develop simple referral criteria, make timely referrals, support patient onboarding, and revisit digital tool use during routine care so technologies extend rather than fragment personalised diabetes support.
Real-world effectiveness and sustainability	Generate longer-term evidence on effectiveness, cost-effectiveness, reach, retention, sustainability, and workforce impact in routine practice.	Embed digital tools into routine pathways with clear ownership, monitoring, and feedback loops so teams can identify low uptake, declining engagement, or unintended workload consequences early.
Theory, mechanisms of action, and fidelity	Use explicit logic models, behaviour change theory, fidelity assessment, and standardised reporting to identify which intervention components work, for whom, under what conditions, and through which mechanisms.	Ensure all team members understand the intended function of digital tools and how to reinforce key components during consultations or follow-up.
Equity and inclusive design	Co-design interventions with underserved groups and evaluate differential access, engagement, and outcomes by age, ethnicity, socioeconomic status, language, health literacy, digital literacy, and co-morbidity.	Offer flexible support, accessible formats, culturally appropriate content, and non-digital or blended alternatives so digital care does not widen existing inequalities.
Engagement and personalisation	Clarify what constitutes meaningful engagement, how engagement changes over time, and how tailoring, reminders, feedback, peer support, and hybrid delivery can support sustained use without increasing burden.	Help patients select digital tools and services aligned with their preferences, confidence, and care goals. Review engagement as part of ongoing care and use brief prompts or follow-up to re-engage people when needed.
Emerging technologies	Assess the acceptability, safety, usability, equity, and clinical value of artificial intelligence, continuous glucose monitoring, wearables, precision medicine approaches, and just-in-time adaptive interventions in diverse T2DM populations.	Maintain human oversight, explain digital outputs clearly, consider data quality and bias, and use emerging technologies to support rather than replace person-centred clinical judgement.
Broader outcomes	Move beyond short-term HbA1c outcomes to include diabetes distress, treatment burden, quality of life, self-efficacy, patient experience, carer impact, and coordination of care.	Use digital tools as adjuncts to holistic care, ensuring psychological, behavioural, social, and clinical needs are considered across primary care, specialist services, dietetics, psychology, pharmacy, and social care.

Abbreviations: T2DM, type 2 diabetes mellitus.

Funding Statement

This review has no external funding. DPF is a National Institute of Health Research (NIHR) Senior Investigator (NIHR203308). The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care.

Abbreviations

ACT, Acceptance and Commitment Therapy; AI, Artificial intelligence; BCT, Behaviour change technique; DSMES, Diabetes Self-Management Education and Support; HCP, Healthcare professionals; JITAIs, Just-in-time adaptive interventions; NHS, National Health Service; RCT, Randomised controlled trial; TIDieR, Template for Intervention Description and Replication; T2DM, Type 2 diabetes mellitus; UK, United Kingdom.

Data Sharing Statement

No datasets were used and/or analysed for the current review article.

Ethics Approval and Informed Consent

As this is a review article, there were no living persons involved in this study. The review is based on data not attributable to individual persons.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

The authors declare no competing interests regarding this work.