Magnitude of A1C improvement in relation to baseline A1C and amount of weight loss in response to intensive lifestyle intervention in real-world diabetes practice: 13 years of observation

How are Machine Learning and Artificial Intelligence Used in Digital Behavior Change Interventions? A Scoping Review

To assess the current real-world applications of machine learning (ML) and artificial intelligence (AI) as functionality of digital behavior change interventions (DBCIs) that influence patient or consumer health behaviors. A scoping review was done across the EMBASE, PsycInfo, PsycNet, PubMed, and Web of Science databases using search terms related to ML/AI, behavioral science, and digital health to find live DBCIs using ML or AI to influence real-world health behaviors in patients or consumers. A total of 32 articles met inclusion criteria. Evidence regarding behavioral domains, target real-world behaviors, and type and purpose of ML and AI used were extracted. The types and quality of research evaluations done on the DBCIs and limitations of the research were also reviewed. Research occurred between October 9, 2023, and January 20, 2024. Twenty-three DBCIs used AI to influence real-world health behaviors. Most common domains were cardiometabolic health (n=5, 21.7%) and lifestyle interventions (n=4, 17.4%). The most common types of ML and AI used were classical ML algorithms (n=10, 43.5%), reinforcement learning (n=8, 34.8%), natural language understanding (n=8, 34.8%), and conversational AI (n=5, 21.7%). Evidence was generally positive, but had limitations such as inability to detect causation, low generalizability, or insufficient study duration to understand long-term outcomes. Despite evidence gaps related to the novelty of the technology, research supports the promise of using AI in DBCIs to manage complex input data and offer personalized, contextualized support for people changing real-world behaviors. Key opportunities are standardizing terminology and improving understanding of what ML and AI are.

A Real-World Precision Medicine Program Including the KidneyIntelX Test Effectively Changes Management Decisions and Outcomes for Patients With Early-Stage Diabetic Kidney Disease

INTRODUCTION/OBJECTIVE

The KidneyIntelX is a multiplex, bioprognostic, immunoassay consisting of 3 plasma biomarkers and clinical variables that uses machine learning to predict a patient's risk for a progressive decline in kidney function over 5 years. We report the 1-year pre- and post-test clinical impact on care management, eGFR slope, and A1C along with engagement of population health clinical pharmacists and patient coordinators to promote a program of sustainable kidney, metabolic, and cardiac health.

METHODS

The KidneyIntelX in vitro prognostic test was previously validated for patients with type 2 diabetes and diabetic kidney disease (DKD) to predict kidney function decline within 5 years was introduced into the RWE study (NCT04802395) across the Health System as part of a population health chronic disease management program from [November 2020 to April 2023]. Pre- and post-test patients with a minimum of 12 months of follow-up post KidneyIntelX were assessed across all aspects of the program.

RESULTS

A total of 5348 patients with DKD had a KidneyIntelX assay. The median age was 68 years old, 52% were female, 27% self-identified as Black, and 89% had hypertension. The median baseline eGFR was 62 ml/min/1.73 m2, urine albumin-creatinine ratio was 54 mg/g, and A1C was 7.3%. The KidneyIntelX risk level was low in 49%, intermediate in 40%, and high in 11% of cases. New prescriptions for SGLT2i, GLP-1 RA, or referral to a specialist were noted in 19%, 33%, and 43% among low-, intermediate-, and high-risk patients, respectively. The median A1C decreased from 8.2% pre-test to 7.5% post-test in the high-risk group (P < .001). UACR levels in the intermediate-risk patients with albuminuria were reduced by 20%, and in a subgroup treated with new scripts for SGLT2i, UACR levels were lowered by approximately 50%. The median eGFR slope improved from -7.08 ml/min/1.73 m2/year to -4.27 ml/min/1.73 m2/year in high-risk patients (P = .0003), -2.65 to -1.04 in intermediate risk, and -3.26 ml/min/1.73 m2/year to +0.45 ml/min/1.73 m2/year in patients with low-risk (P < .001).

CONCLUSIONS

Deployment and risk stratification by KidneyIntelX was associated with an escalation in action taken to optimize cardio-kidney-metabolic health including medications and specialist referrals. Glycemic control and kidney function trajectories improved post-KidneyIntelX testing, with the greatest improvements observed in those scored as high-risk.

Long-term effect of intensive lifestyle intervention on cardiometabolic risk factors and microvascular complications in patients with diabetes in real-world clinical practice: a 10-year longitudinal study

INTRODUCTION

Intensive lifestyle intervention (ILI) has significantly reduced incidence of diabetes and improved many cardiovascular disease risk factors. We evaluated long-term effects of ILI on cardiometabolic risk factors, and microvascular and macrovascular complications among patients with diabetes in real-world clinical practice.

RESEARCH DESIGN AND METHODS

We evaluated 129 patients with diabetes and obesity enrolled in a 12-week translational model of ILI. At 1 year, we divided participants into group A, who maintained <7% weight loss (n=61, 47.7%), and group B, who maintained ≥7% weight loss (n=67, 52.3%). We continued to follow them for 10 years.

RESULTS

The total cohort lost an average of 10.8±4.6 kg (-9.7%) at 12 weeks and maintained an average weight loss of 7.7±10 kg (-6.9%) at 10 years. Group A maintained 4.3±9.5 kg (-4.3%) and group B maintained 10.8±9.3 kg (-9.3%) of weight loss at 10 years (p<0.001 between groups). In group A, A1c decreased from 7.5±1.3% to 6.7±0.9% at 12 weeks but rebounded to 7.7±1.4% at 1 year and 8.0±1.9% at 10 years. In group B, A1c decreased from 7.4±1.2% to 6.4±0.9% at 12 weeks then increased to 6.8±1.2% at 1 year and 7.3±1.5% at 10 years (p<0.05 between groups). Maintenance of ≥7% weight loss at 1 year was associated with a 68% lower risk of developing nephropathy for up to 10 years compared with maintenance of <7% weight loss (adjusted HR for group B: 0.32, 95% CI 0.11, 0.9, p=0.007).

CONCLUSIONS

Weight reduction in patients with diabetes can be maintained for up to 10 years in real-world clinical practice. Sustained weight loss is associated with significantly lower A1c at 10 years and improvement in lipid profile. Maintenance of ≥7% weight loss at 1 year is associated with decreased incidence of diabetic nephropathy at 10 years.