DIA:CONN P8 smart insulin pen
G2E
What it is
A rechargeable, motor-driven smart pen sold in South Korea that delivers U-100 insulin in 0.1-unit steps, records roughly 1,000 injections, and pairs with the DIA:CONN app for CGM-aware bolus calculations. A small, two-week observational study compared 21 P8 users, all with T1D, with 21 controls, including three with T2D. The full mixed cohort favoured P8, but the reported T1D-only subgroup between-group TIR difference was not significant; a larger randomized trial has completed without posted results.
Editorial review: .
Most recent recorded citation date: 2025-09-03. Only explicit date metadata is included; an undated citation may be newer. This does not mean every claim was reviewed on that date.
Trial status, labels and access can change between reviews. How we review the evidence · How to read the evidence
Evidence behind this assessment
Key evidence notes. Study results, product eligibility and access answer different questions.
- Who was studied?
- Study populations and analysis groups vary. Product age limits alone do not describe who was studied.See the linked studies and their populations →
- Benefit or performance
- Dose guidance: The app calculates boluses from clinician-set glucose target, insulin-to- carbohydrate ratio, sensitivity factor, and active-insulin time. That is decision support, not automatic dosing.[4]
- Important harms and treatment burden
- Read the safety discussion and original sources. A missing summary does not establish safety.
- Approval and country access
- Country-specific approval and access are not summarized in this record.Approval, trial recruitment, local supply and funding are separate. Check the cited label or access source.
- Follow-up and remaining uncertainty
- Read the full discussion and original sources for follow-up duration and study limitations.
This record cites company or conference reports. Those sources may describe interim findings; their source type is shown in the source list.
Editorial score: calculation and evidence
A weighted editorial judgment on a 0–100 scale, not a probability of success or a measured treatment effect. Higher criterion scores mean more favorable assessments.
Default calculation: 92 × 18 + 82 × 20 + 78 × 15 + 96 × 12 + 62 × 15 + 42 × 10 + 25 × 10 = 7218; divide by total weight 100. Unrounded weighted result: 72.18.
Stores dose amount and timing for about 1,000 injections and synchronizes them to the DIA:CONN app; it can also record injection site and insulin temperature.[2]
The app calculates boluses from clinician-set glucose target, insulin-to- carbohydrate ratio, sensitivity factor, and active-insulin time. That is decision support, not automatic dosing.[4]
The studied app incorporated current CGM value and trend into dose calculations, but compatibility is app- and market-specific rather than an open CGM interface (site assessment of the G2E system; not stated in the fetched paper text).[4]
Motor-driven dosing is adjustable in 0.1-unit increments, with a stated single-injection range of 0.5 to 60 units.[2]
One-touch injection, OLED display, app synchronization, and rechargeable operation reduce logging work, but the pen is larger than a conventional pen and depends on proprietary refillable cartridges and regular charging (size comparison and burden weightings are site judgment, not sourced measurements).[2]
It supports multiple insulin types through a dedicated 3 mL U-100 cartridge, but remains tied to G2E's pen, app, cartridge, and cloud platform; the Europe-bound P8-11 is a separate not-yet-launched model (platform lock-in framing is site characterisation of the G2E system).[3]
The original P8 has a live Korean purchase and support pathway, but no verified US or European launch. G2E labels the separate P8-11 as launching soon in Europe, without a date.[1]
These tools do not automate basal insulin. Scores reflect what the shipping product does for injection users: reliably capturing doses, turning glucose and insulin-on-board data into useful guidance, and reducing daily burden without hiding regional or insulin-cartridge lock-in.
The full picture
DIA:CONN P8 is not a prototype. G2E's current Korean page links to a purchase site, identifies Huons as the seller, and carries current medical-device advertising approval. The verified commercial footprint is South Korea; the manufacturer's separate P8-11 page says only that a European launch is coming soon. There is no verified US launch.13
The hardware is unusual: a rechargeable motor sets doses in 0.1-unit steps from 0.5 to 60 units, and a dedicated 3 mL cartridge can be filled with supported U-100 insulin. It stores about 1,000 injections and synchronizes them to the app. The app can combine CGM data, carbohydrate entry, clinician-set ratios, and active insulin to calculate a bolus; the person still confirms and injects it. This is a connected injection system, not automated insulin delivery.2
There is early evidence, but the 42-person cohort was not wholly T1D. All 21 P8 users had type 1 diabetes; the 21-person traditional-pen group included 18 people with T1D and three with T2D. After two weeks, TIR rose from 47.8% to 70.8% in the P8 group, and the full mixed-population between-group change favoured P8 by 11.0 percentage points. Allocation was by preference, follow-up was short, and the intervention bundled intensive diabetes education with the pen. Crucially, the between-group TIR difference was not statistically significant in the reported T1D-only subgroup analysis, although time-below-range hypoglycaemia reductions remained significantly greater between groups in that analysis.4
A larger 12-week randomized study has completed, but it mixed type 1, type 2, and post-pancreatectomy diabetes and had posted no results by 8 August 2026. Until those results are available, the manufacturer's feature claims and the small observational signal are the evidence ceiling.5
Sources
- [1]DIA:CONN P8 Korean product and sales-support page · Manufacturer — Identifies G2E as manufacturer and Huons as seller, links to a Korean purchase site, and lists current specifications and medical-device advertising approval. This establishes Korean market status, not clinical benefit.
G2E. DIA:CONN P8 Korean product and support page.
- [2]DIA:CONN P8 product features and specifications · Manufacturer — Manufacturer specifications for 0.1-unit adjustment, 0.5-to-60-unit injections, 3 mL cartridge, motor drive, app functions, and dose history.
G2E. DIA:CONN P8 features and specifications (page served only a generic shell in-session; hardware wording corroborated via the Korean and global pages).
- [3]DIA:CONN global product page · Manufacturer — Distinguishes the commercially documented Korean P8 from the P8-11, which is described only as launching soon in Europe.
G2E. DIA:CONN global product page.
- [4]Effectiveness of the Stage 4 Smart Insulin Pen DIA:CONN P8 for Glycemic Control in a Real-World Setting · Peer-reviewed study · 2025-09-03 — Single-centre retrospective comparison of 42 adults with intensively treated diabetes: all 21 P8 users had T1D, while the traditional-pen group included 18 people with T1D and three with T2D. Follow-up was two weeks; allocation was not randomized. TIR rose from 47.8% to 70.8% in the P8 group, and the full-cohort between-group TIR change was 11.0 percentage points (p=0.046). In the T1D-only comparison, that TIR difference was not statistically significant, although hypoglycaemia reductions remained significant.
Kwon SY, et al. Effectiveness of the Stage 4 Smart Insulin Pen DIA:CONN P8 for Glycemic Control in a Real-World Setting. Diabetes Metab J. 2026 (volume, pages and PMCID as cited; paper substance verified via the journal page).
- [5]DIA:CONN P8 randomized smart-pen study (NCT06406439) · Trial registry — Completed 12-week, 120-participant study across type 1, type 2, and post-pancreatectomy diabetes. No results were posted when checked on 8 August 2026, so it does not yet establish a T1D benefit.
ClinicalTrials.gov. DIA:CONN P8 (Smart Insulin Pen) Study in Diabetes Patients (NCT06406439).