GRAPHICAL ABSTRACT

Highlights

• First Korean real-world study evaluating stage 4 smart insulin pens (SIPs).

• SIP provides CGM-integrated bolus calculator with insulin-on-board tracking.

• SIP users showed significantly better TIR, TBR, and GRI than traditional pens.

• Improvements attributed to SIP’s precise dosing via bolus calculation.

• SIP may enhance glycemic control beyond traditional pen use in insulin-treated DM.

INTRODUCTION

Continuous glucose monitoring (CGM) and automated insulin delivery (AID) systems have improved glucose control; however, barriers, such as cost and the need for continuous pump use, limit their accessibility [1,2]. The stage 4 smart insulin pen (SIP), which integrates several functions of AID systems, including a bolus calculator and integration with CGM, provides a cost-effective and accessible alternative for individuals with intensively insulin-treated diabetes [1-3]. Despite its potential benefits, trials and real-world data on this device are limited.

This study aimed to retrospectively analyze real-world data from individuals with insulin-treated diabetes who used the DIA:CONN P8 (G2E, Seoul, Korea), the first commercially available SIP in Korea, to evaluate its efficacy in improving glucose control in clinical practice.

METHODS

Study design

Between January 2024 and July 2024, all adult patients (>18 years) with insulin-treated diabetes who underwent intensive diabetes self-management education (DSME; median 3.8 hours) at the Daegu Catholic University Medical Center were sequentially collected in this retrospective observational study. The inclusion criteria comprised the use of basal-bolus insulin therapy and real-time CGM (Dexcom G6/G7 [Dexcom, San Diego, CA, USA] or CareSense Air [i-SENSE, Seoul, Korea]) for >3 months. Among these individuals, only those with suboptimal glycemic control, defined as an glycosylated hemoglobin (HbA1c) >7.0%, a percentage of time in range (%TIR) <70%, or a percentage of time below range (%TBR) <70 mg/dL exceeding 4% during the baseline period (2 weeks prior to DSME), were included. To minimize selection bias, all eligible patients who presented at the clinic during the study period (n=49) were included sequentially. Exclusion criteria included pregnancy (n=2), inability to understand DSME (n=4), or insufficient use of the bolus calculator (<70% of bolus doses; n=4) in the SIP group.

A total of 42 patients were included in the final analysis. After completing DSME (median 4.2 hours for the traditional insulin pen [TIP] group, 3.4 hours for the SIP group), which included training on carbohydrate counting, CGM data interpretation, insulin-to-carbohydrate ratios (ICR), insulin sensitivity factors (ISF), and hypoglycemia management, 21 patients transitioned to SIP use based on personal preference, whereas 21 continued using TIPs. In the TIP group, patients self-adjusted insulin doses based on their DSME. In the SIP group, dose adjustments were guided by the bolus calculator, and patients were instructed to follow its recommendations. The study protocol was approved by the Institutional Review Board of Daegu Catholic University Hospital (no. 2025-02-024), and the requirement for informed consent was waived because the data were anonymized.

Smart insulin pen DIA:CONN P8

The DIA:CONN P8 is a pen-shaped insulin delivery device connected to a smartphone application that provides bolus insulin calculations based on customizable settings, including target glucose, ICR, ISF, and active insulin time. The application integrates CGM data to automatically incorporate current glucose levels and trends into dose calculations. The application offers bolus calculation modes suited to user expertise: fixed meal dose settings for predefined insulin doses or flexible calculations for manual carbohydrate input. All SIP users in this study were encouraged to use the flexible calculation method.

Both groups used disposable insulin pens for basal insulin (insulin degludec or insulin glargine U300), with SIPs used exclusively for bolus injections in the SIP group.

Data collection and analysis

Glycemic metrics were collected from CGM readings, including the %TIR 70–180 mg/dL, percentage of time above range (%TAR) >180 and >250 mg/dL, %TBR <70 and <54 mg/dL, coefficient of variation (CV), and glucose management indicator (GMI). The glycemic risk index (GRI), which assesses glycemic quality, was calculated based on CGM parameters [4].

These metrics were assessed during the baseline period (2 weeks prior to DSME) and the 2-week periods of SIP use (SIP group) or continued TIP use (TIP group) following DSME (Supplementary Fig. 1). Clinical variables, including C-peptide and HbA1c levels, were obtained from the most recent measurements within 1 year prior to baseline CGM data collection.

The primary outcome was the mean difference in glycemic metrics before and after DSME in the TIP group compared with the mean difference before DSME and after SIP initiation in the SIP group, assessed using Mann-Whitney U tests. Within-group changes in glycemic metrics were analyzed using Wilcoxon signed-rank tests.

RESULTS

Baseline characteristics and glycemic outcomes in each group

All the participants in the SIP group had type 1 diabetes mellitus (T1DM), whereas the TIP group included three patients with type 2 diabetes mellitus (T2DM). No significant differences were observed in the baseline characteristics, except for %TBR <54 mg/dL (Supplementary Table 1).

Fig. 1, Supplementary Table 2 show the glycemic changes between the baseline and follow-up periods in the TIP and SIP groups. In the TIP group, %TIR, %TAR >250 mg/dL, %TAR >180 mg/dL, GMI, and GRI were significantly improved (40.9%–53.0%, P=0.005; 28.2%–20.0%, P=0.014; 58.1%–44.9%, P=0.005; 8.4–7.7, P=0.004; 71.5–57.3, P=0.006, respectively). However, %CV showed no significant change (36.4%–34.7%, P=0.263), and %TBR worsened, although the changes were not significant (%TBR <70 mg/dL: 1.0%–2.1%, P=0.101; %TBR <54 mg/dL: 0.1%–0.4%, P=0.157). In the SIP group, %TIR, %TAR >250 mg/dL, %TAR >180 mg/dL, %TBR <70 mg/dL, %TBR <54 mg/dL, GMI, and GRI were significantly improved (47.8%–70.8%, P<0.001; 21.3%–7.2%, P<0.001; 49.2%–27.7%, P<0.001; 3.1%–1.5%, P=0.032; 0.7%–0.2%, P=0.046; 7.9–7.0, P<0.001; and 64.2–31.7, P<0.001, respectively). The %CV decreased from 36.5% to 34.2% (P=0.103); however, this change was not significant. In both groups, total daily dose of insulin (TDD) decreased numerically but did not reach statistical significance.

Comparison of glycemic outcome between the SIP and TIP groups

Fig. 2, Supplementary Table 3 present the differences in the glycemic outcomes between the SIP and TIP groups. The differences between the groups in %TIR, %TBR <70 mg/dL, %TBR <54 mg/dL, and GRI were significantly greater in the SIP group (11.0%, P=0.046; –2.6%, P=0.024; –0.9%, P=0.027; –18.2, P=0.022, respectively). Although the differences between the groups in %TAR, CV, GMI, and TDD were not statistically significant, the SIP group showed numerically better improvements in these parameters compared with the TIP group. A subgroup analysis restricted to patients with T1DM, excluding the three individuals with T2DM in the TIP group, was conducted (Supplementary Table 4). Although the between-group difference in %TIR was not statistically significant, the SIP group still demonstrated numerically greater improvement, and the reductions in %TBR remained significantly greater in the SIP group.

DISCUSSION

This study retrospectively analyzed real-world data from patients with insulin-treated diabetes who began using the SIP to evaluate its efficacy in glycemic control, compared with those who maintained their usual TIP. Patients who received DSME showed significant improvements in %TIR, %TAR, GMI, and GRI, regardless of whether they continued using TIPs or started using SIPs. However, %TBR tended to worsen numerically in TIP users, whereas SIP users demonstrated improvements. When comparing changes in CGM metrics between the two groups, SIP use resulted in superior improvements in %TIR, %TBR, and GRI compared with maintaining TIP. Although not statistically significant, SIP users showed numerically better trends across other CGM parameters. These findings suggest that, even under similar clinical conditions, such as receiving DSME and using CGM, switching to SIP is associated with superior glycemic outcomes.

Similar to previous studies on SIP [5-9], this study demonstrated significant improvements in glycemic outcomes. In an observational study of Swedish adults using SIP, significant improvements were observed in CGM parameters [6]. Another observational study in children with T1DM also reported significant improvements in %TBR [5]. In our study, the SIP also showed significant improvements across all CGM parameters. Furthermore, unlike the TIP group, the SIP group exhibited improvements in %TBR, suggesting the potential benefits of SIP in reducing the risk of hypoglycemia. However, it should be noted that the baseline %TBR in the SIP group was significantly higher than in the TIP group, which may have influenced these findings.

Intensive DSME is associated with improved glycemic control in patients with diabetes [10]. However, the observed improvements in glycemic control in the SIP group could not be attributed solely to DSME. Notably, SIP users received less DSME time on average than TIP users. Additionally, all SIP users had T1DM, whereas the TIP group included patients with T2DM and exhibited higher mean C-peptide levels, which likely conferred a relative advantage in glycemic control [11,12]. The superior outcomes in the SIP group may be attributed to the bolus calculator, which enables more precise insulin dosing and helps prevent postprandial hyperglycemia. The calculator also incorporates insulin-on-board tracking, which helps prevent insulin stacking and thus reduces the risk of hypoglycemia. In our dataset, there were no significant changes in TDD before and after SIP use, suggesting that the improved glycemic outcomes were not due to increased insulin administration but rather to enhanced dosing precision facilitated by SIP functions [1-3,13].

To our knowledge, this is the first study to evaluate the real-world use of stage 4 SIPs in Korea. The study not only compared data before and after SIP use, but also included a comparison with TIP users who received the same DSME, allowing for a clearer assessment of SIP’s intrinsic effects. However, this study was subject to selection bias due to its single-institution setting and the inclusion of highly motivated patients. The relatively small sample size and short SIP application period limited the ability to observe long-term effects. An ongoing multicenter randomized controlled trial (NCT06406439) assessing SIP use is expected to validate these findings.

In conclusion, this analysis of real-world SIP and CGM data suggests that SIP use is associated with improved glycemic management in adults with intensively insulin-treated diabetes.

SUPPLEMENTARY MATERIALS

NOTES

CONFLICTS OF INTEREST

No potential conflict of interest relevant to this article was reported.

AUTHOR CONTRIBUTIONS

Conception or design: S.Y.K., J.H.K.

Acquisition, analysis, or interpretation of data: S.Y.K., H.K.

Drafting the work or revising: S.Y.K.

Final approval of the manuscript: J.H.K.

FUNDING

None

ACKNOWLEDGMENTS

None

Fig. 1.

Changes in glycemic outcomes in the (A) traditional insulin pen and (B) smart insulin pen groups. Bars represent pre- and post-intervention glycemic metrics. Mean changes with 95% confidence intervals (CIs) for each variable are presented on the right. TAR 250, percentage of time above range >250 mg/dL; TAR 180, percentage of time above range >180 mg/dL; TIR, percentage of time in range (70–180 mg/dL); TBR 70, percentage of time below range <70 mg/dL; TBR 54, percentage of time below range < 54 mg/dL; GMI, glucose management indicator; CV, coefficient of variation; GRI, glycemic risk index. Footnotes indicate statistical significance based on within-group comparisons using the Wilcoxon signed-rank test: ^aP<0.05, ^bP<0.01, ^cP<0.001.

Fig. 2.

Comparison of glycemic outcome changes between the traditional insulin pen (TIP) and smart insulin pen (SIP) groups. (A) Percentage of time above range (TAR) >250 mg/dL, (B) percentage of TAR >180 mg/dL, (C) percentage of time in range (TIR), (D) percentage of time below range (TBR) <70 mg/dL, (E) percentage of TBR <54 mg/dL, (F) coefficient of variation (CV), (G) glucose management indicator (GMI), (H) glycemic risk index (GRI). The bar graphs display the changes in glycemic outcomes, including GMI, GRI, %TIR, %TAR, %TBR, and CV, between the baseline and follow-up periods in the TIP (orange) and SIP (blue) groups. Between-group differences and corresponding P values are presented for each glycemic metric.

REFERENCES

• 1. Kompala T, Neinstein AB. Smart insulin pens: advancing digital transformation and a connected diabetes care ecosystem. J Diabetes Sci Technol 2022;16:596-604.ArticlePubMedPDF
• 2. Yoo JH, Kim JH. Advances in continuous glucose monitoring and integrated devices for management of diabetes with insulin-based therapy: improvement in glycemic control. Diabetes Metab J 2023;47:27-41.ArticlePubMedPMCPDF
• 3. MacLeod J, Vigersky RA. A review of precision insulin management with smart insulin pens: opening up the digital door to people on insulin injection therapy. J Diabetes Sci Technol 2023;17:283-9.ArticlePubMedPDF
• 4. Klonoff DC, Wang J, Rodbard D, Kohn MA, Li C, Liepmann D, et al. A glycemia risk index (GRI) of hypoglycemia and hyperglycemia for continuous glucose monitoring validated by clinician ratings. J Diabetes Sci Technol 2023;17:1226-42.PubMed
• 5. Adolfsson P, Bjornsson V, Hartvig NV, Kaas A, Moller JB, Ogionwo Lange E. Improved glycemic control observed in children with type 1 diabetes following the introduction of smart insulin pens: a real-world study. Diabetes Ther 2022;13:43-56.ArticlePubMedPDF
• 6. Adolfsson P, Hartvig NV, Kaas A, Moller JB, Hellman J. Increased time in range and fewer missed bolus injections after introduction of a smart connected insulin pen. Diabetes Technol Ther 2020;22:709-18.ArticlePubMedPMC
• 7. Danne TP, Joubert M, Hartvig NV, Kaas A, Knudsen NN, Mader JK. Association between treatment adherence and continuous glucose monitoring outcomes in people with diabetes using smart insulin pens in a real-world setting. Diabetes Care 2024;47:995-1003.ArticlePubMedPMCPDF
• 8. Gomez-Peralta F, Abreu C, Fernandez-Rubio E, Cotovad L, Pujante P, Gaztambide S, et al. Efficacy of a connected insulin pen cap in people with noncontrolled type 1 diabetes: a multicenter randomized clinical trial. Diabetes Care 2023;46:206-8.ArticlePubMedPDF
• 9. MacLeod J, Im GH, Smith M, Vigersky RA. Shining the spotlight on multiple daily insulin therapy: real-world evidence of the InPen smart insulin pen. Diabetes Technol Ther 2024;26:33-9.ArticlePubMedPMC
• 10. Kim JY, Jin SM, Sim KH, Kim BY, Cho JH, Moon JS, et al. Continuous glucose monitoring with structured education in adults with type 2 diabetes managed by multiple daily insulin injections: a multicentre randomised controlled trial. Diabetologia 2024;67:1223-34.ArticlePubMedPDF
• 11. Christensen MB, Gaede P, Hommel E, Gotfredsen A, Norgaard K. Glycaemic variability and hypoglycaemia are associated with C-peptide levels in insulin-treated type 2 diabetes. Diabetes Metab 2020;46:61-5.ArticlePubMed
• 12. Gibb FW, McKnight JA, Clarke C, Strachan MW. Preserved C-peptide secretion is associated with fewer low-glucose events and lower glucose variability on flash glucose monitoring in adults with type 1 diabetes. Diabetologia 2020;63:906-14.ArticlePubMedPMCPDF
• 13. Castle JR, Wilson LM, Tyler NS, Espinoza AZ, Mosquera-Lopez CM, Kushner T, et al. Assessment of a decision support system for adults with type 1 diabetes on multiple daily insulin injections. Diabetes Technol Ther 2022;24:892-7.ArticlePubMedPMC

Figure & Data

References

Citations

Citations to this article as recorded by  

About this article

Kwon SY, Kwak H, Kim JH. Effectiveness of the Stage 4 Smart Insulin Pen DIA:CONN P8 for Glycemic Control in a Real-World Setting. Diabetes Metab J. 2025 Sep 3. doi: 10.4093/dmj.2025.0112. Epub ahead of print.

Received: Feb 11, 2025; Accepted: Mar 23, 2025

DOI: https://doi.org/10.4093/dmj.2025.0112.

Download citation ↓