Two‐centre real‐world pilot of a biopsychosocial digital health tool to deliver personalised visits and enhance value of rtCGM

Relentless diabetes management and high rates of diabetes distress persist, leading to negative long-term consequences.1 Psychological support is rarely available even though psychological morbidity is known to worsen diabetes outcomes by impairing effective self-management and contributes significantly to personal and societal costs.2 Despite the many advances in diabetes devices and pharmacological treatments, widespread suboptimal glycaemia remains.3 There is a clear and urgent need for interventions that go beyond the physical model of health care, to routinely include psychological and social well-being.4 Such considerations are an important missing component in a health system poorly suited to supporting optimal diabetes outcomes.4 Health care professional (HCP) burnout is at an all-time high as clinicians struggle with unprecedented numbers of patients, restricted resources and ongoing widespread disruption to clinical services.5 An inability to signpost or deliver psychosocial resources currently leaves clinicians at a disadvantage. Devices associated with improved glycaemic outcomes are often associated with unacceptable burdens for the user resulting in cessation of use.6 A survey of 127 clinicians6 found that nervousness about reliance on technology (40%) or that the device might not work (26%) were commonly cited barriers to uptake by their patients. Additional barriers included interference with sleep (51%), alarms (75%) and insufficient time during clinic visits to learn about how to use the device (15%). To address these barriers and create a more holistic approach to diabetes care, we developed the Spotlight-AQ online tool and conducted a two-centre, real-world pilot study (NHS ethics approval IRAS: 315237). Participants were a convenience sample of adults with type 2 diabetes (T2D) aged ≥18 years and were using injectable therapy to treat their diabetes. All participants were provided real-time continuous glucose monitoring (rtCGM). Spotlight-AQ uses a digital health platform to facilitate the delivery of biopsychosocial health care as part of routine outpatient diabetes appointments. The intervention comprises a short questionnaire (3–5 minutes) sent to participants automatically with a visit reminder, a week prior to their appointment. Questions are algorithmically calculated and address psychological, social, skills and therapy-related aspects of diabetes to identify priority concerns of the participant. Results are immediately visible to the patient via smartphone, tablet or computer in an easy-to-understand diagram. Clinicians’ secure portal login facilitates viewing results of each participant, with mapped, evidence-based psychosocial resources, and enables tailored and personalised discussions within the visit. Designed for routine 15- and 30-minute visits, the intervention has been shown to significantly reduce consultation length due to more focussed conversations and reduce HCP burnout.7 We assessed glycated haemoglobin (HbA1c), change in weight, diabetes distress and glucose monitoring satisfaction at baseline and at three months. Associations between baseline measures and demographic factors were examined. SPSS version 20.0 was used to conduct statistical analyses via repeated-measures analysis of variance, t test and correlations. Between-group comparisons, adjustment for baseline variables was carried out using HbA1c, time in range, Diabetes Distress Scale and Glucose Monitoring System Satisfaction Survey as outcome variables. Twenty-three participants were recruited between 30 August 2022 and 30 November 2022. Two participants dropped out of the study, leaving 21 participants, of whom eight were women; mean age of participants was 64 years (range 48–86). See Table 1. Female Male 5 6 4 8 White British Chinese 12 - 11 1 Once-weekly GLP1 injectable Daily injectable GLP1 Fixed daily insulin doses 5 1 11 1 - 11 Overall mean HbA1c fell by 1.4% (15mmol/mol), range: 4-71mmol/mol, p≤0.001). The reduction in HbA1c was greater at centre two compared with centre one (28mmol/mol [2.4%] vs 8mmol/mol [0.7%]). HbA1c improved in 15 participants, remained unchanged in one participant and increased for five participants. There was no difference in the change in HbA1c between men (16mmol/mol; 1.6%) and women (14mmol/mol; 1.3%). Mean DDS total scores show participants reporting moderate diabetes distress at baseline. Subscale scores demonstrate that interpersonal distress and emotional burden both reflected moderate distress, with regimen distress showing high distress. Follow-up scores demonstrate significant improvements in the total score and all subscales. GMSS scores did not improve overall or within worthwhileness or openness subscales. Emotional burden and behavioural burden were significantly reduced. Therefore, subgroup analyses were not conducted. This two-centre pilot study shows that the use of a digital health tool, in conjunction with the initiation of rtCGM, was associated with a significant improvement in HbA1c levels and reduced diabetes distress for people with T2D. Furthermore, weight loss was achieved by half of the participants. It has long been recognised that a biopsychosocial model of health care is necessary to support people living with diabetes. Such a model that incorporates physical, psychological and social well-being is consistent with WHO guidelines and regulatory approval bodies including the Food and Drug Administration and Medicines and Healthcare Products Regulatory Agency. A major challenge to the delivery of such health care is the requirement for transformation from the medical model currently employed in most health care systems globally. Increasingly, person-reported outcomes have been introduced to varying degrees in routine care; however, this has not always taken an evidence-based approach and has often fallen short of desired benefits. Our intervention is designed for use within routine care visits to reduce burden on HCPs and improve health outcomes for people with diabetes. It helps focus visit discussions, aiding understanding of patient-identified priorities. By doing so, it provides an additional tool for HCPs to tailor their support to the unique needs of each patient. Psychosocial factors implicated in diabetes management, including emotional distress, well-being and coping, are all modifiable. A recent randomised controlled trial (RCT) evaluation of the current intervention demonstrated both physical and mental health improvements for adults with T1D and T2D.7 Widespread dissatisfaction with the lack of psychosocial support by people with diabetes is significantly related to self-care behaviours, resulting in poorer quality of life and worse blood glucose control.8 It was encouraging to observe weight reduction in more than half of participants, particularly as T2D is often characterised by overweight and obesity. Improvement of blood glucose levels and an associated weight loss can significantly reduce an individual's risk of developing complications, including cardiovascular events, and improve quality of life.9 Although it could be argued that the results of the study could be attributable to the rtCGM intervention, intervention clearly conferred greater benefits than rtCGM alone. The systematic review by Kieu et al.10 explored the benefits of utilising CGM in primary care, reviewing 10 studies (n=4006 participants) and found CGM was more effective at reducing HbA1c levels than usual care (−0.43%), with low evidence of improved time in target range. Our results demonstrate a considerable added benefit alongside rtCGM. Our mean HbA1c improvement of 15mmol/mol (range 4–71mmol/mol) was significantly better than that reported by Kieu et al. This cumulative benefit demonstrates the strength of adopting a biopsychosocial approach within routine care alongside devices. The strengths of the study include the delivery of biopsychosocial support alongside rtCGM to demonstrate a significant cumulative benefit. These data are consistent with, and build on, previous feasibility and RCT7 data. The study is limited by its small size and lack of control group. The increased interaction with HCPs could have influenced the results. In conclusion, the intervention was associated with significant improvement in glycaemic levels and significantly reduced diabetes-related distress beyond rtCGM alone. We would like to thank all of our participants for their willingness to take part. This study was supported by an unrestricted research grant from Dexcom. R.K. and K.B.K. are founders and shareholders of Spotlight-AQ Ltd.