What the research found

Researchers tracked over 56,000 people with type 2 diabetes using a remote monitoring program between 2021 and 2025. They found that estimated A1c (calculated from fasting glucose readings via connected meters) declined gradually—by about 0.06 percentage points per year on average. Even after accounting for medications, treatment adherence, demographics, and app engagement, this improvement held up, declining about 0.04 points yearly.

The team then broke down what drove these improvements. Adoption of newer cardiorenal-protective drugs (GLP-1 agonists and SGLT-2 inhibitors) accounted for the largest share of benefit. Interestingly, while greater engagement with the app and more frequent glucose monitoring were linked to better outcomes, the overall program engagement metrics actually partially offset some of the glycemic gains—suggesting these patients may have been monitored more closely because they had poorer baseline control. Notably, half of the improvement couldn't be explained by any measured factors, hinting at unmeasured lifestyle changes or other influences. Over the five-year window, the proportion of participants hitting an A1c below 7% rose from about 75% to 80%.

Why it matters for you

If you're a MyKine user tracking glucose and considering GLP-1 or SGLT-2 inhibitor use, this is relevant context: the data suggests these drugs do contribute meaningfully to long-term A1c reduction in a real-world setting, not just in trials. The improvement was steady and durable, not a flash-in-the-pan effect that wore off.

That said, the finding that roughly half the benefit was unexplained is equally instructive. This likely reflects sustained dietary habits, training adherence, sleep consistency, or stress management—the inputs you're already tracking in MyKine. The implication is that medication is part of the picture, but the measurable lifestyle factors still matter. Frequent glucose monitoring itself correlated with better control, even if causality is unclear; it may simply keep you honest or alert to patterns.

For users on TRT or peptide protocols, this study doesn't directly address androgens or other compounds, so don't over-extrapolate. But the methodological point stands: remote, continuous biomarker tracking plus structured engagement appears to sustain improvements over years, not months.

Caveats

• Observational and non-randomized: No control group; impossible to isolate cause from correlation, especially for engagement metrics
• Industry-affiliated: Data come from Teladoc Health, a commercial diabetes management platform; outcomes reflect their patient population and protocols
• Half the effect unexplained: The authors themselves flag that demographic, treatment, and engagement factors explain only ~49% of the trend, limiting insight into actual drivers
• A1c estimates, not measured: eA1c was calculated from fasting glucose, not direct A1c lab draws, so absolute values should be interpreted cautiously
• Survivorship and dropout bias possible: A 5-year cohort inevitably loses participants; unclear how non-adherence or attrition shaped results