Abstract
Background:
Type 1 diabetes mellitus (T1DM) management requires intensive insulin therapy to maintain glycemic control. Basal-bolus injections (multiple daily injections, MDI) and continuous subcutaneous insulin infusion (CSII) via insulin pumps are commonly used regimens. In resource-limited settings, the choice of therapy is influenced by factors such as cost, accessibility, and effectiveness. Limited data exist comparing these modalities among Indian patients with T1DM.
Aims:
To compare the effectiveness of insulin pump therapy versus basal-bolus insulin injections in Indian patients with T1DM using continuous glucose monitoring system (CGMS) data.
Methodology:
Fifty-seven individuals with T1DM monitored using CGMS were included and divided into two groups: 31 on basal-bolus therapy (Group A) and 26 on insulin pump therapy (Group B). Mean ages were 18.9 ± 5.0 years (Group A) and 16.4 ± 5.1 years (Group B). Baseline data, including demographics, duration of diabetes, insulin dose, and HbA1c, were collected. Glycemic metrics from CGMS—time in range (TIR), time above range (TAR), and time below range (TBR)—were analyzed.
Results:
Baseline characteristics were similar between groups, with no significant differences in body mass index, diabetes duration, or initial HbA1c. The insulin pump group required significantly lower insulin doses (24.4 ± 7.9 units vs. 43.5 ± 15.8 units; P < 0.0001). Mean TIR was slightly higher in the MDI group (38.9% vs. 32.9%; P = 0.1806), although more patients in the pump group achieved target TIR >70% (7.7% vs. 0%). TAR was higher in the pump group (55.7% vs. 43.5%; P = 0.0522), while TBR was lower (11.3% vs. 20.0%; P = 0.0662). A greater proportion of the pump group achieved the target TBR <4% (46.2% vs. 19.3%; P = 0.03).
Conclusions:
In this resource-limited setting, insulin pump therapy was associated with lower insulin requirements and better TBR outcomes, although overall glycemic control (TIR and TAR) was comparable between CSII and MDI. Both therapies can be effective, and optimal outcomes depend on individualized care and consistent glucose monitoring.
Introduction
Type 1 diabetes mellitus (T1DM), accounting for 5%–10% of all cases of diabetes mellitus, requires lifelong insulin treatment to optimize blood glucose levels to prevent diabetic ketoacidosis and vascular complications. 1 Common regimens include multiple daily injections (MDI) of insulin analogues and continuous subcutaneous insulin infusion (CSII) via insulin pumps. Multiple daily subcutaneous injections help sustain the background insulin levels (basal insulin) and address the increased insulin levels needed to manage postprandial hyperglycemia (bolus insulin)(1). It has less equipment with minimum maintenance and is easy to use. Furthermore, it is cost-effective compared with insulin pumps. A CSII or insulin pump involves using a small electronic device to deliver a continuous supply of insulin (basal or bolus) into the body throughout the day, mimicking pancreatic function. 2 The use of CSII has been associated with a reduced risk of hypoglycemia, improper dosing, and avoiding multiple injections a day. 3 However, it may be associated with pump malfunction, skin problems, and difficulties in performing physical activity. 4
Another factor hindering its adoption in T1DM management is the high cost of pumps, costing up to $5185, including consumables such as cannulas, reservoirs, and batteries. 1 A cost-effectiveness analysis in India observed that Insulin pumps cost about $15,297 to gain an additional year of perfect health (DALY), 5 while the national average cost-per-DALY is $997 ($667 to $1316). 6 Furthermore, due to limited coverage and accessibility of health care insurance, out-of-pocket expenses account for 60% of all health care transactions. 7 Therefore, in resource-limited settings, the choice between these therapies depends on efficacy. There are limited data on the comparative effectiveness of these treatment modalities in the Indian setting. Two studies conducted in the Indian T1DM population concluded comparable efficacy of the two treatment modalities.2,4 The current study aims to compare the effectiveness of insulin pump therapy versus basal-bolus insulin injections among Indian patients with T1DM in a resource-limited setting using continuous glucose monitoring system (CGMS) data to assess glycemic control.
Methods
Study design and setting
This prospective observational study was conducted over a period of 1 year from January 2024 to December 2024 at the type 1 diabetes specialty clinic of Department of Endocrinology at a tertiary care center. The study aimed to evaluate the glycemic efficacy of different insulin delivery modalities—MDI using a basal-bolus regimen versus CSII—in children with T1DM, using data obtained from CGMS. The insulin pumps used (MiniMed 700 G, Medtronic Ltd.) were provided free of cost by the state government health program, thereby facilitating access for eligible patients despite financial constraints.
Participants
Children aged 3 to 18 years with a confirmed diagnosis of T1DM were eligible for inclusion. Participants were on a stable insulin regimen (either basal-bolus or insulin pump therapy) for at least 3 months prior to enrollment and were under regular follow-up at the study center. Informed consent was obtained from parents or guardians, and assent was obtained from children, as appropriate. The study protocol was reviewed and approved by the Institutional Ethics Committee of Nizam’s Institute of Medical Sciences (EC/NIMS/3623/2023).
Inclusion criteria
Children aged 3–18 years with T1DM.
Receiving insulin therapy via either a basal-bolus regimen or insulin pump for at least 3 months.
Under regular follow-up at the study center.
Willingness to participate with informed consent from parents/guardians.
Exclusion criteria
Presence of conditions affecting glucose metabolism (e.g., acute and chronic infections, hospitalization).
Recent change in insulin regimen within the last 3 months.
Incomplete or poor-quality CGMS data (<70% sensor wear time during the 14-day monitoring period).
Known allergy or skin intolerance to CGMS sensor adhesives.
Inability to adhere to study protocols or follow-up schedule.
Data collection and monitoring
All enrolled participants underwent CGMS using the FreeStyle Libre Pro system. Sensors were applied at scheduled intervals for 14 days during the course of the study. Data were downloaded and analyzed for glycemic parameters including time in range (TIR, 70–180 mg/dL), time below range (TBR, <70 mg/dL), time above range (TAR, >180 mg/dL), mean glucose levels, and glycemic variability. The data were collated and entered into an Excel sheet. Demographic details such as age, body mass index (BMI), and duration of diabetes were obtained during the study’s initiation. The dosage of insulin used was also recorded. Adverse events such as diabetic ketoacidosis and patterns of hyperglycemic events were recorded.
CGMS data were stratified based on the mode of insulin delivery (MDI vs. CSII) and analyzed to assess comparative efficacy in achieving glycemic targets. Sensor data were only included for analysis if ≥70% of readings were available during the 14-day period.
Statistical analysis
The data were analyzed and representations were developed using GraphPad PRISM software. Categorical variables were expressed as proportions, N (%) and continuous variables were expressed as mean ± SD.
Results
A total of 57 individuals with T1DM were included in the study and monitored using CGMS. Participants were divided into two groups: 31 patients receiving basal-bolus insulin therapy (Group A) and 26 patients on insulin pump therapy (Group B).
Baseline characteristics
Most participants were adolescents and young adults. The mean age was 18.9 ± 5.0 years in Group A and 16.4 ± 5.1 years in Group B. No statistically significant differences were observed between the groups in terms of BMI (17.3 ± 4.1 vs. 17.5 ± 3.3; P = 0.8421), duration of diabetes (4.1 ± 3.6 vs. 5.9 ± 3.7 years; P = 0.0687), or baseline HbA1c levels (10.4% ± 3.1% vs. 11.5 ± 1.6%; P = 0.1079).
Insulin dose
Group B required a significantly lower total daily insulin dose than Group A (24.4 ± 7.9 units vs. 43.5 ± 15.8 units; P < 0.0001).
Time in range
The average TIR (70–180 mg/dL) was 38.9% ± 14.7% in Group A and 32.9% ± 18.7% in Group B (P = 0.1806). A higher proportion of Group B achieved the target TIR (>70%) compared with Group A (7.7% vs. 0%).
Time above range
Group B showed a higher average TAR (>180 mg/dL) of 55.7% ± 23.5% compared with 43.5% ± 22.8% in Group A (P = 0.0522). Less than 20% of patients in either group achieved the target TAR.
Time below range
The TBR (<70 mg/dL) was lower in Group B (11.3% ± 15.1%) compared with Group A (20.0% ± 19.2%; P = 0.0662). A greater proportion of patients in Group B achieved the target TBR (<4%) (46.2% vs. 19.3%).
Acute complications
During the study period, there were four documented episodes of diabetic ketoacidosis in three children on MDI, whereas there were none in the insulin pump group. There were two episodes of hospitalization for hypoglycemia in the MDI group and none in the insulin pump group.
The results have been summarized in Table 1.
Comparison of Physical, Clinical and Glycemic Characteristics Between T1DM Patients on Basal-Bolus Injections and Continuous Insulin Infusion
Group A, MDI; Group B, insulin pump; MDI, multiple daily injections; DKA, diabetic ketoacidosis.
Discussion
India is home to over 250,000 individuals with T1DM. 7 The use of insulin pumps in India is often limited by high costs, insufficient infrastructure, and a shortage of trained health care personnel. Particularly in rural areas, inconsistent government supplies of insulin and limited access to diabetes technology make pump therapy a challenge.8,9 This study was designed to evaluate whether basal-bolus insulin therapy can offer comparable glycemic outcomes to insulin pump therapy in such resource-constrained settings.
Our findings reveal that while insulin pump therapy (CSII) significantly reduced total daily insulin requirements, it did not result in statistically superior glycemic control compared with MDI in terms of TIR, TAR, or TBR. However, a higher proportion of pump users achieved target TBR, possibly due to the better basal coverage of insulin throughout the day, taking into account each child’s individual insulin needs. The TBR was lower in the insulin pump group (11.3% ± 15.1%) compared with the basal-bolus group (20.0% ± 19.2%; P = 0.0662). Although this mean difference did not reach statistical significance, interpretation in the context of T1DM is important. Since patients typically spend a relatively small proportion of TBR, even modest absolute differences in TBR may hold clinical relevance in terms of hypoglycemia risk. Higher generation pumps have features like threshold-suspend in sensor-augmented pump systems, which help mitigate hypoglycemia risk. 10 Although Group B (CSII) had a higher average TAR, the difference was borderline significant (P = 0.0522), and fewer patients in both groups achieved the recommended targets for TIR and TAR. This is likely reflective of the limited adherence and variable education levels in the study population. Previous studies have reported higher success rates with insulin pumps in terms of TIR and TAR,11–13 underscoring the importance of support systems and patient engagement in achieving optimal outcomes.
Interestingly, the greater proportion of insulin pump users achieving target TBR supports evidence from global studies indicating CSII’s effectiveness in reducing hypoglycemia, even at lower insulin doses. 14 In CSII, the basal dose of insulin was higher, and the bolus dose was lower. This also resulted in a reduction in the total daily basal insulin dose. 9 The Henry activity curve for insulin suggests that the higher the concentration of premeal insulin, the greater its effect on reducing nonmeal blood glucose and the stronger the “tail effect” on premeal glucose levels. 14 This mechanism may help explain the numerically lower insulin requirements observed in the CSII group, although overall glycemic outcomes did not differ significantly between groups.
An important observation from this study is the critical role of self-monitoring of blood glucose and timely adjustment of insulin doses in achieving optimal glycemic outcomes. Independent of the insulin delivery modality, patients who monitored their glucose more consistently tended to have better glycemic metrics, particularly higher TIR. This underscores that patient engagement, education, and empowerment to act on glucose trends are as essential as the choice of treatment modality itself.
Although the insulin pumps used in this study (MiniMed 700G, Medtronic Ltd.) are basic models without full automation or advanced hybrid closed-loop features, they still demonstrated efficacy when paired with appropriate monitoring and titration strategies. One of the significant benefits observed was the reduction in hypoglycemia, likely due to the programmable basal rate and bolus flexibility offered by pumps, even in their simpler versions. In resource-limited settings, where access to advanced diabetes technology is restricted, basic insulin pumps—if supported by structured education, glucose monitoring, and clinical follow-up—can serve as a cost-effective and valuable tool in managing T1DM in children and adolescents.
Conclusions
T1DM requires lifelong insulin treatment. The study suggests that while insulin pump therapy may reduce insulin dosage and achieve a higher proportion of individuals meeting the recommended TBR target, its overall glycemic efficacy was comparable with basal–bolus therapy in this resource-limited setting. Depending on individual patient factors and resource availability, both treatment modalities remain viable for managing T1DM. These findings should be interpreted in light of the study’s limited power, and larger, adequately powered studies are warranted to further clarify the potential advantages of CSII.
Limitations
The study’s small sample size and lack of adherence tracking limit the generalizability of these findings. Adherence and regular self-monitoring of glucose play a crucial role in glycemic outcomes, and their omission may confound the comparative efficacy of treatment modalities. Larger, prospective studies including adherence monitoring are warranted to evaluate the true effectiveness of insulin delivery systems in real-world settings. Another limitation is the lack of systematically collected data on adherence and structured diabetes education. Given their known impact on glycemic outcomes, future studies should incorporate standardized assessment of these factors to better contextualize treatment effects.
Footnotes
Authors’ Contributions
B.A. contributed to the conceptualization, writing—original draft, writing—review and editing; B.K. contributed to data curation and formal analysis; S.R.K. contributed to supervision, validation, writing—review and editing; G.T. contributed to data curation and methodology; N.D. contributed to data curation and validation; and B.S.G. contributed to formal analysis, methodology and validation.
Disclosure Statement
The authors have nothing to disclose.
Funding Information
No funds received from any source for this study.