Insulin pump therapy (CSII – Continuous Subcutaneous Insulin Infusion) is the gold standard of insulin treatment, primarily for patients with type 1 diabetes mellitus, and in rare cases, for patients with other types of diabetes as well.
Who can use an insulin pump?
Not every person with type 1 diabetes wants or needs an insulin pump for their therapy.
The rate varies greatly by country and healthcare access:
Currently, in Germany, around 140,000 people with type 1 diabetes use such a device, which represents about 38% of the roughly 370,000 affected individuals [1]. Even a “classic” insulin pump therapy is expensive: the pump itself costs €3,000–4,000 and is typically used for four years, which corresponds to about €3 per day. Annual costs for consumables such as infusion sets, catheters, etc., amount to another €1,000–2,000, adding roughly €5 per day. Therefore, the daily therapy cost is about €8 higher than for intensified conventional therapy (ICT).
October 2025 overview of estimated insulin pump costs and annual infusion set/catheter expenses across Europe, the USA, and India.
(Figures are approximate and vary by model, manufacturer, insurance coverage, and local pricing.)
Region |
Pump purchase cost |
Annual cost for consumables (infusion sets, catheters, reservoirs, etc.) |
Europe |
Example (UK): about £2,000–£3,000 (~€2,300–€3,500) per pump. (BreakthroughT1D UK) |
European studies estimate around €1,000–€1,400 per year for infusion sets and reservoirs. (Wikipedia – Insulin pump) |
USA |
Without insurance: typically US$4,500–8,000+ per pump. (ProMed DME) |
Consumables: roughly US$1,500–4,000 per year. (ProMed DME) One Medicare-based estimate (including pump, supplies, insulin) was US$3,329/year amortized. (CMS.gov) |
India |
Typical retail price ₹100,000–₹600,000 (~€1,100–€6,600), depending on model and brand. (HospitalStore.com) |
Example pricing: infusion set (pack of 10) ₹3,797, reservoir (pack of 10) ₹1,485. (T1DF India) |
Notes & limitations
When a CGM (Continuous Glucose Monitoring) system is added to perform sensor-augmented pump therapy (SuP) or to use a hybrid closed-loop / automated insulin delivery (AID) system, additional costs for glucose sensors arise. Depending on the system, these range between €1,700 and €4,200 per year (the wide price range is due to sensor wear time, which varies between 7 and 15 days).
Thus, AID therapy in Germany about €15–20 more per day than ICT.
Approximate costs for continuous glucose monitoring (CGM) sensors in Europe, the USA and Andia. These are rough estimates, since actual out-of-pocket costs vary widely depending on insurance, country reimbursement, brand/model and usage.
USA
Summary for USA: Expect anything from ~US$2,000-6,000/year (or US$160-500/month) before insurance.
Europe
Estimate for Europe: Based on available data, CGM sensors might cost roughly €250-300/month (≈ €3,000-3,600/year) before reimbursement in many Western European countries—but actual costs could be lower or higher depending on country subsidies.
India
Summary for India: The cost is highly variable. Some data suggest ₹5,000+ per sensor unit, but actual monthly costs for using a CGM may reach ₹10,000-20,000/month (~₹120,000-₹240,000/year) in some cases.
Important caveats
Country-specific pricing examples for continuous glucose monitoring (CGM) sensors in Germany, the UK, India (as available, 2025/10), along with some product listings. Please note: These are retail or indicative costs and may not reflect reimbursed amounts, insurance coverage, or bulk/contract pricing.
Germany
Interpretation: In Germany, brush-prices for sensors can be in the range of ~70-100 € each (or more depending on model). Yearly costs for someone using a sensor every ~2 weeks could sum to ~1.500-2.000 € or higher (depending on frequency and model).
United Kingdom
Interpretation: In the UK, some sensors are priced quite lower (in the budget ~£23-£40 range) under certain procurement/contracted contexts, but retail/online may show higher prices (~£75). For regular replacement (say every 14 days), annual cost might around £500-£900+ depending on model and usage.
India
Interpretation: In India, pricing is very variable. One sensor item shows ~₹5,000, but in practice recurring costs (especially with imported devices, frequent replacement, no reimbursement) may lead to many tens of thousands of rupees per year.
Summary Table (Estimated)
Country |
Approx sensor / unit cost |
Annual estimate (device only) |
Germany |
~70-100 € each |
~1.500-2.000 € or higher |
UK |
~£23-£75 each |
~£500-£950+ |
India |
~₹5,000+ each |
₹60,000+ to ₹240,000+ or more |
Important Caveats
The daily therapy costs are higher than for intensified conventional therapy (ICT).
However, the investment pays off: most AID users achieve HbA1c values around 7%, a Time in Range (TIR) of 70% or higher, and only a low risk of severe hypoglycemia [2,3]. This means that diabetic complications become unlikely — at least if AID therapy is started early.
The following section is only a rough overview of reimbursement eligibility in each region and is subject to change
United States
Regulatory status
Who qualifies (coverage/eligibility)
Common restrictions
Europe (EU/EEA + UK examples)
Regulatory status
Who qualifies (examples of national rules)
Common restrictions across Europe
India
Regulatory status
Who qualifies (coverage/eligibility)
Common restrictions
In brief:
How are pumps regulated?
Are there restrictions?
Yes—everywhere. Pumps are prescription devices, usually require specialist initiation and training, and payer criteria (e.g., documented hypoglycaemia, failure of optimized MDI, or specific lab evidence in the US). Device-specific age indications also apply. (NICE)
Who is entitled to a pump?
It can also be observed that there is greater openness to prescribing insulin pumps for children and adolescents with diabetes. Because of the effectiveness of CSII combined with an AID system (improving metabolic control and reducing acute complications), international guidelines recommend AID as a first-line therapy for children and adolescents with type 1 diabetes [6]. In Germany, about 70–80% of these young people with diabetes use such a system.
For AID systems, approval is in some cases somewhat easier due to the typically excellent glucose control they achieve.
Fig. T1D 6
Change in HbA1c after switching patients with type 1 diabetes from ICT with regular/NPH insulin to ICT with a rapid-acting insulin analogue/glargine, and after a second switch to CSII (left). On the right, the change in HbA1c is shown in relation to the initial HbA1c value. [10]
This individual example is confirmed by meta-analyses.
An analysis of 22 randomized controlled trials comparing CSII with ICT demonstrated that, on average, CSII improves HbA1c by 0.55% [14].
In another meta-analysis, the results of CSII therapy in patients with type 1 diabetes were examined with respect to the frequency of severe hypoglycemia [15]. Again, 22 studies (10 of them involving children and adolescents) met the strict criteria regarding study duration, patient numbers, etc. The analysis showed that all studies reported a significant reduction in the number of severe hypoglycemic events. Overall, the rate of severe hypoglycemia decreased by a factor of 4.19—that is, by more than three quarters to 23.9%. At the same time, HbA1c improved by an average of 0.61%.
Compared with ICT, CSII reduces the risk of developing cardiovascular disease and lowers mortality [16]. This was shown in an analysis of the Swedish National Diabetes Register, which includes data on about 95% of all people with type 1 diabetes in Sweden.
When using CSII, the event rate per 1,000 patient-years was lower compared with ICT (ICT = 1.0):
In addition to these significant outcomes, CSII improves the quality of life of people with diabetes. It allows for a more flexible daily routine, much less restricted by insulin therapy. This leads to greater resilience and performance. Irregular professional duties—such as those of business travelers or physicians—can be managed more easily, as can various leisure activities [17].
The basal rate, i.e., the distribution of insulin delivery over 24 hours, is often not constant but follows a circadian rhythm.
In the early morning hours, insulin requirements are at their highest due to increased insulin resistance, then remain constant throughout the day, and slightly rise again in the evening.
The distribution roughly corresponds to the basal rate pattern shown below.
The circadian pattern of insulin requirements changes over a person’s lifetime. In long-standing diabetes, the circadian distribution gradually approaches a more constant insulin demand over the full 24-hour period.
In general, less insulin is required under CSII than under previous ICT using basal (intermediate-acting) insulin. This also applies to the basal rate, i.e. the insulin needed independently of food intake.
The determination of the basal rate depends largely on the glycemic control achieved during ICT. If glycemic control was satisfactory, with only a few hypoglycemic episodes, the basal rate is reduced to about 85% of the previous total.
If, on the other hand, hypoglycemia occurred frequently, a reduction of about 30% is recommended.
Thus, it is generally appropriate to initially reduce the basal insulin dose by around 30% compared to ICT.
Confirm readiness & choose the system
Wait-out method: Start full programmed basal when glargine is essentially gone (~24 h after the last dose).
Overlap-safe method: Start the pump sooner but reduce/zero the basal for 6–24 h (e.g., temp basal 0–70%) to account for residual glargine, then return to the programmed basal when the long-acting insulin has worn off. Manufacturer order sets explicitly recommend a temporary basal of 0% until the last long-acting dose is inactive; ADA pump-start texts note 30–50% basal reduction for 6–24 h when residual basal is present.
both methods are acceptable in the U.S.
Patient scenario
Method A – “Conservative pump-start” approach
(Used by many U.S. centers to reduce early hypoglycemia risk)
Time block |
Relative adjustment |
Rate (U/h) |
Notes |
00:00–03:00 |
90 % of avg |
0.55 |
prevent nocturnal lows |
03:00–07:00 |
120 % of avg |
0.70 |
dawn phenomenon |
07:00–12:00 |
90 % of avg |
0.55 |
high morning sensitivity |
12:00–18:00 |
100 % of avg |
0.60 |
baseline |
18:00–24:00 |
110 % of avg |
0.65 |
evening rise |
Mean |
— |
≈ 0.61 U/h → ≈ 15 U/day |
Start in manual mode (not full AID) for 24 h.
Perform fasting-block basal tests over the next week and adjust each segment ±10–20 %.
Method B – “Glargine ÷ 24” approach
(Often used if the clinician wants to match the prior basal exposure)
Time block |
Relative adjustment |
Rate (U/h) |
Notes |
00:00–03:00 |
90 % of avg |
0.75 |
prevent nocturnal lows |
03:00–07:00 |
120 % of avg |
1.00 |
dawn rise |
07:00–12:00 |
90 % of avg |
0.75 |
high sensitivity |
12:00–18:00 |
100 % of avg |
0.83 |
baseline |
18:00–24:00 |
110 % of avg |
0.90 |
evening tendency to rise |
Mean |
— |
≈ 0.83 U/h → ≈ 20 U/day |
matches prior glargine |
Because residual glargine may still be active on day 1, most U.S. clinics start this profile at 50–70 % temp basal for 6–24 h, then return to full rates once glargine has worn off.
After pump start
The Methods often fit quite well and can be fine-tuned using basal rate tests. However, basal rate testing is now only rarely necessary, since insulin pumps are mostly used together with CGM systems. This combination makes it easy to adjust the basal rate, and hypoglycemia—especially at night—can be detected early.
With the large number of daily glucose profiles available, it is usually possible to determine whether an issue is due to an incorrect basal rate or an inadequate bolus dose. Therefore, skipping meals to test the basal rate (the “meal omission test”) is now often unnecessary.
However, in some cases, this approach is still useful for precisely determining the correct basal rate.
In an AID system, the glucose sensor controls the basal insulin delivery of the pump. The algorithm automatically adjusts the initially programmed basal rate to the patient’s actual needs (as with the Tandem t:slim X2 with Control-IQ) or determines it dynamically at short intervals (e.g., every five minutes with Medtronic MiniMed insulin pumps).
The Temporary Basal Rate
In CSII or sensor-augmented pump therapy (SuP), the basal rate can be temporarily increased or decreased as needed. This allows for adjustments to changes in insulin requirements—for example, a reduction during physical activity or an increase during acute illness.
In AID systems, these adjustments occur automatically via CGM measurements and the pump’s insulin delivery algorithm.
In general, the basal rate can be raised or lowered to match the individual’s insulin needs. Thus, changes in insulin requirements due to exercise or illness can be managed dynamically.
In closed-loop systems, the basal rate is automatically adapted based on continuous glucose monitoring.
An insulin pump also offers the advantage of variable bolus delivery. The bolus can be administered as a standard bolus, a dual bolus, or an extended (square-wave) bolus.
For slowly absorbed carbohydrates, such as those in fat-rich meals, a delayed bolus can be used—where part of the bolus insulin is delivered immediately and the remainder over a defined period, e.g. two to three hours.
Modern insulin pumps also include a bolus calculator.
Using predefined therapy settings, the calculator suggests an insulin bolus based on the measured blood glucose level and intended carbohydrate intake.
It also takes into account the duration of action of a previous bolus, helping to prevent
In a modern AID system, the insulin pump is informed only about the carbohydrate amount, or the user simply indicates the type or size of the meal (e.g., small, medium, large; breakfast, lunch, dinner).
These parameters must be defined before the first use of the AID system.
The AID algorithm then calculates the insulin bolus and compensates for fluctuations in insulin requirements by adjusting the basal insulin delivery accordingly.
For most CGM systems, routine fingerstick calibration is no longer necessary, although it can still be performed (except with the FreeStyle Libre system, which does not allow calibration).
Occasional blood glucose measurements, however, are still recommended to verify the accuracy of the system.
Fig. T1D 7:
Insulin action profile during insulin pump therapy.
Only rapid-acting insulin is infused: the individually programmed basal rate covers the insulin requirement independent of food intake, while boluses are delivered at mealtimes or to correct elevated glucose levels.
(Blue dashed lines: insulin levels in people without diabetes.)
Communicating with the Computer
All insulin pumps are equipped with an intelligent data management system that, via appropriate interfaces, allows therapy data and basal rate settings to be read and analyzed on an external computer.
There are also mobile applications (apps) available for use on smartphones.
[1] Pankowska E et al.: Three kind of mealtime insulin administration and metabolic control in diabetic children on insulin pumps. Diabetologia 2008; 51 (Suppl. 1), S381.
[2] Pańkowska E, Błazik M, Groele L. Does the fat-protein meal increase postprandial glucose level in type 1 diabetes patients on insulin pump: the conclusion of a randomized study. Diabetes Technol Ther. 2012 Jan;14(1):16-22.
[3] Sherr JL, Schoelwer M, Dos Santos TJ, et.al.. ISPAD Clinical Practice Consensus Guidelines 2022: Diabetes technologies: Insulin delivery. Pediatr Diabetes. 2022 Dec;23(8):1406-1431. doi: 10.1111/pedi.13421. Epub 2022 Dec 5. PMID: 36468192.
[4] Castañeda J, Arrieta A, van den Heuvel T, et.al. Time in Tight Glucose Range in Type 1 Diabetes: Predictive Factors and Achievable Targets in Real-World Users of the MiniMed 780G System. Diabetes Care. 2024 May 1;47(5):790-797.
[5] Silva JD, Lepore G, Battelino T, et.al.. Real-World Performance of the MiniMed™ 780G System: First Report of Outcomes from 4120 Users. Diabetes Technol Ther. 2022;24(2):113-119.
[6] von dem Berge T, Remus K, Biester S, et.al. In-home use of a hybrid closed loop achieves time-in-range targets in preschoolers and school children: Results from a randomized, controlled, crossover trial. Diabetes Obes Metab. 2022 Jul;24(7):1319-1327.
[7] Petry F. AID-System sorgt für überraschende Wende im Diabetesmanagement. In|Fo|Diabetologie 2024;18 (6):6-9.
[8] Heise T, Nosek L, Rønn BB, et.al. Lower within-subject variability of insulin detemir in comparison to NPH insulin and insulin glargine in people with type 1 diabetes. Diabetes. 2004;53(6):1614-1620.
[9] Datenbank Pubmed, Stichwort “diabetes csii”. Letzter Aufruf: 02.06.2025.
[1013] Pickup JC, Kidd J, Burmiston D et al. Effectiveness of continuous subcutaneous insulin infusion in hypoglycaemia-prone type 1 diabetes. Practical Diabetes International 2005; 22: 10 – 14.
[14] Jeitler K, Horvath K, Berghold A et al. Continuous subcutaneous insulin infusion versus multiple daily insulin injections in patients with diabetes mellitus: systematic review and meta-analysis. Diabetologia 2008; 51: 941-951.
[15] Pickup J, Sutten AJ Severe hypoglycaemia and glycaemic control in Type 1 diabetes: meta-analysis of multiple daily insulin injections compared with continuous subcutaneous insulin infusion. Diabetic Medicine 2008; 25: 765 – 774.
[16] Steineck I, Cederholm J, Eliasson B, et.al; Swedish National Diabetes Register. Insulin pump therapy, multiple daily injections, and cardiovascular mortality in 18,168 people with type 1 diabetes: observational study. BMJ. 2015 Jun 22;350:h3234.
[17] Liebl A, Krinelke L Long-term results of insulin pump therapy (CSII) in adolescents and adults with type 1 diabetes mellitus in Germany. Diabetologia Polska 2003; 10: 179-185.