Diabetes technology and monitoring

• Monitoring turns diabetes from a guess into a measurement. HbA1c shows the average over about 2 to 3 months, a blood glucose meter gives a point reading, and continuous glucose monitoring (CGM) streams interstitial glucose so that time spent in target range can be seen directly. ^[1]
• For many nonpregnant adults the A1C goal is under 7 percent (under 53 mmol/mol), with a time-in-range goal above 70 percent on CGM. Stricter or more lenient goals are chosen by health, hypoglycemia risk, age and life expectancy ^[1].
• Devices should be offered to people with diabetes, including CGM from diagnosis when insulin is used, and insulin pumps or automated insulin delivery (AID) systems are the preferred insulin delivery method for type 1 diabetes ^[2].
• CGM does not remove the need for finger-stick checks: capillary blood glucose is still used to confirm unexpected readings and when the sensor is warming up or not transmitting ^[2].
• NICE recommends that adults with type 1 diabetes are offered a choice of real-time or intermittently scanned CGM, and aims them at an HbA1c of 48 mmol/mol (6.5 percent) or lower where hypoglycemia can be avoided ^[3].

By the end of this page you should be able to:

1. State the common HbA1c goals and explain when HbA1c is unreliable.
2. Describe how a blood glucose meter is used well, including when to check and how often.
3. Interpret a standardized CGM report: wear time, mean glucose, coefficient of variation, time above, in and below range.
4. Set CGM targets for typical adults and for older adults with complex health.
5. Explain the indications, practical benefits and limitations of insulin pumps and AID systems.
6. Recognize device-specific cautions, including interfering substances and skin reactions.
7. Perform monitoring safely in pregnancy, childhood, older age, kidney disease and hemoglobin disorders, and in low-resource settings.

Diabetes technology and monitoring is a procedure: it is the repeatable clinical task of obtaining glucose information, setting goals, choosing devices and delivery systems, and acting on the data. This page covers the common monitoring tools, HbA1c and its limits, blood glucose meters, continuous glucose monitoring, the meaning of time in range, and the hardware that delivers insulin, including pumps and automated insulin delivery systems. Drug treatment of diabetes itself is left to the relevant pharmacology topics; here the emphasis is on measurement and safe use of devices.

The guidance quoted is the newest open edition available at the time: the American Diabetes Association (ADA) Standards of Care in Diabetes—2026, sections 6 and 7 ^[1][2], and the National Institute for Health and Care Excellence (NICE) guideline NG17 for type 1 diabetes in adults, last updated 17 August 2022 ^[3].

Monitoring means two different jobs. The first is looking back, to judge long-term control: that is what HbA1c does. The second is looking now and ahead, to steer doses and prevent harm: that is what blood glucose monitoring and CGM do. Technology adds the hardware and software that can adjust insulin delivery automatically, notably pumps and AID systems ^[2].

A diabetes device should be offered to people with diabetes ^[2]. The type and selection should be individualized for the person's needs, circumstances, preferences and skill level, and for children or for adults with cognitive or physical limitations the caregiver's skills are part of the decision ^[2].

• Anyone with diabetes needs glycemic assessment at least twice a year, and more often—typically every 3 months—when goals are not met, treatment has recently changed, hypoglycemia or hyperglycemia is frequent or severe, health status changes, or a child is growing rapidly ^[1].
• People taking insulin need monitoring to detect and prevent hypoglycemia and hyperglycemia, and to tune doses.
• CGM is particularly useful in people at risk of hypoglycemia and in type 1 diabetes, and its use is growing in type 2 diabetes and other forms of diabetes, especially when insulin is used ^[1].
• Monitoring is also used when an HbA1c result disagrees with the clinical picture, when symptoms suggest hypoglycemia, during illness, pregnancy, exercise, driving and work with safety implications, and when medication is changed.

• Automated insulin delivery is the preferred insulin delivery method over multiple daily injections (MDI), conventional pump therapy and sensor-augmented pumps in type 1 diabetes, in adults and in adolescents with type 2 diabetes treated with insulin, and in other insulin-deficient forms of diabetes. It can be considered for selected people with type 2 diabetes on basal insulin who are not at their individualized goals ^[2].
• Early initiation of CGM and pump or AID therapy should be considered, including from diagnosis, based on the person's or caregiver's needs and preferences ^[2].
• In the UK, hybrid closed loop systems are recommended for adults with type 1 diabetes whose HbA1c is 58 mmol/mol (7.5 percent) or more, or who have disabling hypoglycemia despite best possible management with a pump, real-time CGM or intermittently scanned CGM ^[3].

There are few absolute contraindications to glucose monitoring, but device choice has safety conditions.

• People with diabetes who use CGM must still have access to a blood glucose meter at all times, because every CGM system can be wrong, and because meters are needed for confirmation, calibration when required, during rapid change and in clinical settings ^[2].
• CGM should not be treated as accurate during warm-up, transmission gaps, severe hypoglycemia where symptoms disagree with the reading, or when interfering substances are present ^[2] (see Table 3).
• Insulin pumps and AID systems require hands-on training and ongoing support for safe use; they should not be started with education alone and with no follow-up ^[2].
• C-peptide level, islet autoantibodies and duration of insulin treatment should not be required before starting pump or AID therapy; none predicts responsiveness ^[2].
• More intensive goals are inappropriate when the risk of hypoglycemia, cognitive or functional limitation, frailty or comorbidity is high ^[1].

HbA1c reflects average glycemia over approximately 2 to 3 months. It is an indirect measure: factors that alter hemoglobin or red blood cells (hemolysis, anemia, sickle cell disease and other homozygous hemoglobin variants, glucose-6-phosphate dehydrogenase deficiency, recent transfusion, erythropoiesis-stimulating drugs, kidney failure, pregnancy) make it invalid. In these situations glycemic status is assessed with CGM, blood glucose monitoring or glycated serum protein assays such as fructosamine ^[1][3].

Capillary blood glucose is a point measurement of plasma glucose. CGM measures interstitial glucose, which correlates well with plasma glucose but lags when glucose is rising or falling rapidly ^[2]. Because the readings are continuous, CGM can display not only the average but also variability, the time below, in and above target, and events that a single finger-stick or an HbA1c can miss.

Approved meters with proven accuracy should be used, with unexpired strips from a pharmacy or licensed distributor, stored properly. Meters differ in accuracy; in one comparison 14 of 18 meters met the minimum accuracy requirements ^[2]. People should be advised against preowned or secondhand strips, which can give wrong results ^[2].

Meter accuracy standards differ by setting. The FDA home-use standard requires 95 percent of results within 15 percent for all blood glucose values in the usable range, and 99 percent within 20 percent. The ISO 15197:2013 standard requires 95 percent within 15 percent for blood glucose of 100 mg/dL or more and within 15 mg/dL below 100 mg/dL, plus 99 percent in zones A or B of the consensus error grid ^[2]. Many people assume a meter is accurate simply because it is cleared; this is not guaranteed.

Meter readings can be falsely affected by oxygen tension, temperature and interfering substances including acetaminophen, vitamin C, dopamine, hydroxyurea, N-acetylcysteine, maltose, galactose, xylose, pralidoxime, vitamin C, high or low hematocrit and high altitude ^[2].

Personal CGM systems are of two broad categories: prescribed real-time systems with alarms, and over-the-counter systems without alarms or alerts for people whose treatment does not cause hypoglycemia. Professional systems owned by the clinic are applied for short periods to look at patterns ^[2].

CGM measures interstitial glucose and does not always match capillary glucose when levels are changing quickly. Its accuracy is affected by substances listed in Table 3, and contact dermatitis—both irritant and allergic, sometimes linked to isobornyl acrylate—occurs with all devices that attach to the skin ^[2].

Table 3. Substances that can make a CGM sensor read higher than the true glucose level ^[2]:

• Multiple daily injections using pens is the baseline. Insulin pens are preferred over vials and syringes for most people on MDI, but syringes remain acceptable according to preference, dose, availability and cost. Insulin pens can be used for 14 to 56 days once started, needle gauges range from 22 to 34, and needles 4 to 5 mm long reduce intramuscular injection and possible lipohypertrophy ^[2].
• Connected insulin pens and pen caps can record and transmit doses, pair with some CGM systems and produce downloadable reports. On average, missing as little as two basal doses or four bolus doses in 14 days is associated with a clinically relevant fall of 5 percent or more in time in range ^[2].
• Modern pumps deliver rapid-acting insulin continuously through tubing and a cannula, or through a tubeless pod attached to the skin, and both forms are licensed for type 1 and type 2 diabetes ^[2].
• AID systems add an algorithm that modulates or replaces basal delivery every 5 minutes using CGM data. Targets can be fixed or adjustable from 87 to 180 mg/dL (4.8 to 10 mmol/L), and doses still need meal entry; physical-activity modes are available on most systems ^[2].
• Standalone pumps have been available for over 40 years. In a systematic review and meta-analysis pump therapy in type 1 diabetes lowered A1C by about 0.30 percent and reduced severe hypoglycemia in children, adolescents and adults compared with multiple daily injections ^[2].

• Choose the device and delivery system by shared decision-making, taking needs, circumstances, preferences and skill level into account. For children or adults with cognitive or physical limitations the caregiver's skills matter as much as the patient's ^[2].
• Plan initial and ongoing education before starting a CGM, and specific training before starting a pump or AID system. CGM training can sometimes be self-directed with manufacturer materials; pump and AID systems usually need a certified diabetes educator or trained specialist ^[2]. Schools and workplaces need backup plans and reasonable accommodations ^[2].
• Confirm that supplies, sensors, strips, cannulas and batteries can be afforded and replaced when needed; interrupted access worsens outcomes ^[2].
• Obtain informed consent for any skin sensor or infusion set, and discuss skin reactions, scarred sites and dermatitis before starting.
• Document the negotiated HbA1c goal, CGM goals and treatment plan in the care plan; in the UK record the proportion reaching the audit target ^[3].

For many nonpregnant adults the HbA1c goal is under 7 percent (under 53 mmol/mol), with CGM time in range above 70 percent. For healthy, low-risk, long-life-expectancy individuals a lower HbA1c (under 6.5 percent or under 48 mmol/mol) may be appropriate. For frail, significantly impaired or high-hypoglycemia-risk adults a higher goal, up to 8 percent (64 mmol/mol) or less stringent individualized goals, is appropriate. NICE aims to support adults with type 1 diabetes to 48 mmol/mol (6.5 percent) or lower where hypoglycemia is not problematic, and audits the proportion reaching 53 mmol/mol (7 percent) or lower ^[1][3].

Goals should be agreed jointly with the person and reassessed over time, and more aggressive goals should be modified when treatment causes hypoglycemia ^[1].

• Measure HbA1c routinely for all people with diabetes at initial assessment and in continuing care. Every 3 months judges whether goals are reached and maintained; adults with type 1 or type 2 diabetes who have reached and are maintaining their goal range often need testing only twice a year. NICE recommends testing every 3 to 6 months in type 1 diabetes ^[1][3].
• HbA1c is reported to NGSP standards and reflects 2 to 3 months of glycemia. Laboratory assays are preferred over point-of-care testing, which can be less accurate ^[1].
• Explain the result and keep the most recent HbA1c available at consultations ^[3].

Use CGM or blood glucose profiles, and where hemoglobin turnover is disturbed or hemoglobin variants are present use fructosamine (or glycated albumin), quality-controlled blood glucose profiles or total glycated hemoglobin ^[1][3]. This is mandatory in sickle cell disease and other homozygous hemoglobin variants, where A1C cannot be measured at all ^[1]. Race and ethnicity should not determine how HbA1c is used clinically ^[1].

Most people on intensive insulin therapy (MDI or pump) need to check 6 to 10 times daily if they rely on a meter alone: before meals and snacks, at bedtime, occasionally after meals, before, during and after exercise, when hypoglycemia or hyperglycemia is suspected, after treating hypoglycemia until normoglycemia, and before and during safety-critical tasks such as driving ^[2]. Checking more often is associated with lower A1C, about 0.2 percent per extra check daily in children and adolescents with type 1 diabetes ^[2].

For people on basal insulin only, or on oral and noninsulin injectable therapy, the evidence is weaker, but fasting measurements guide dose adjustment and lower A1C. For noninsulin-treated type 2 diabetes, benefit is most consistent when monitoring is part of a structured program that translates results into dose, diet or activity changes. NICE advises routine self-monitoring at least 4 times daily in type 1 diabetes, including before each meal and bedtime, up to 10 times daily when goals are not reached, hypoglycemia is frequent, during illness, sport, driving, pregnancy and breastfeeding, and more than 10 times daily for lifestyle or impaired awareness reasons ^[2][3].

Technique matters: check before eating, use approved meters and unexpired strips, wash and dry the finger, obtain an adequate sample, avoid secondhand strips, and know the meter's interfering substances. If a reading does not fit the clinical picture, retest or use a laboratory value.

Capillary targets from NICE for type 1 diabetes are fasting 5 to 7 mmol/L on waking, 4 to 7 mmol/L before other meals, and 5 to 9 mmol/L at least 90 minutes after eating. Bedtime targets should be agreed individually and fit the fasting target ^[3].

Start CGM early in type 1 and type 2 diabetes treated with insulin, and consider it for people on noninsulin therapy that can cause hypoglycemia, or wherever it helps management. CGM used as close to daily as possible gives the most benefit, and uninterrupted supply of sensors prevents the deterioration seen when CGM is stopped ^[1][2]. NICE offers adults with type 1 diabetes a choice of real-time or intermittently scanned CGM, with selection based on accuracy, alarms and sharing, dexterity, skin sensitivity, body image, calibration needs and pump integration, and in cost terms when more than one device suits the person equally ^[3].

Professional or intermittent CGM is an option when daily CGM is not feasible, to guide dose and lifestyle change, and should always be coupled with review and education ^[2].

Among the randomized data: CGM reduces A1C and hypoglycemia when sensors are worn regularly. In insulin-treated type 2 diabetes it increases time in range 70 to 180 mg/dL (3.9 to 10 mmol/L) and TBR but not level 3 hypoglycemia rates. Removing CGM from people with type 2 diabetes on basal insulin partially reverses the improvements, so continued use matters ^[2].

A standardized report (single page at minimum, such as the standardized CGM report with weekly summary) should be used for review by both person and clinician ^[2].

Table 1 shows the CGM goals from the ADA Standards of Care, 2026, adapted from the international CGM consensus. For valid assessment the CGM should provide a 10- to 14-day assessment, with at least 70 percent of time active (about 10 of 14 days) ^[1].

Notes:

• The coefficient-of-variation target of 36 percent or lower is for limiting hypoglycemia risk; some studies suggest that 33 percent or lower may give additional protection. ^[1]
• In older adults with complex or intermediate health the aim is 50 percent (12 hours a day) in range and no more than 1 percent (15 minutes a day) below 70 mg/dL, with up to 50 percent of the day above target if needed to prevent hypoglycemia. ^[1]
• Time in range above 70 percent aligns with an HbA1c of about 7 percent (about 53 mmol/mol), although the GMI calculated from CGM can differ from laboratory HbA1c ^[1].

CGM is also used to define hypoglycemia: a clinically important level 1 episode is glucose under 70 mg/dL (under 3.9 mmol/L), level 2 is glucose under 54 mg/dL (under 3.0 mmol/L), and level 3 is severe, with altered mental or physical function requiring another person's help, regardless of the meter reading ^[1]. A reading of 70 mg/dL is the adrenergic threshold in people without diabetes ^[1].

Indications and benefits:

• AID is the preferred delivery method in type 1 diabetes and in adults and adolescents with insulin-treated type 2 diabetes, and can be considered for selected people on basal insulin who are not at goal ^[2].
• In youth with newly diagnosed type 1 diabetes, starting an AID system within 21 days of diagnosis produced time in range 10 percent higher and lower A1C at 12 months than usual care ^[2].
• In the UK, insulin pump therapy is an option per NICE technology appraisal TA151, and hybrid closed loop is an option for adults with HbA1c of 58 mmol/mol (7.5 percent) or higher, or disabling hypoglycemia despite best management with pump or CGM ^[3].

Practical use:

• Start at diagnosis or any time later; these devices can be used safely in adults, older adults, adolescents and children. There is no consensus on which factors predict success, but understanding troubleshooting is mandatory ^[2].
• Complications include infusion-set dislodgement or occlusion leading to ketosis or diabetic ketoacidosis, and skin problems including lipohypertrophy and infection ^[2]. These must be recognized early.
• Pumps and AID should be offered without requiring C-peptide, islet autoantibodies or minimum diabetes duration ^[2]. Access should continue regardless of age or A1C ^[2].

• Review CGM and pump data at each visit and remotely when possible; standardized reports and regular downloads support timely dose changes ^[2].
• Because real-world data show that missing even two basal or four bolus doses in 14 days lowers time in range by about 5 percent, connected pens and downloads are used to spot adherence problems ^[2].
• Hypoglycemia review occurs at every visit for at-risk people, with screening for impaired hypoglycemia awareness at least annually and referral for structured education if awareness is reduced ^[1].

• Level 1 hypoglycemia is treated in a conscious person with fast-acting carbohydrate, avoiding fat- and protein-rich foods, repeated after 15 minutes if it persists ^[1].
• Glucagon should be prescribed for everyone taking insulin or at high risk of hypoglycemia, with family or caregivers trained and ready-to-use preparations preferred ^[1].
• Deintensify insulin, sulfonylureas or meglitinides, or switch to a lower-risk class, when hypoglycemia goals are not met or when treatment burdens outweigh benefits, especially in vulnerable adults ^[1].
• CGM is recommended for people at high hypoglycemia risk ^[1].

• Review sensor insertion sites and skin regularly; contact dermatitis can occur with any skin-adherent device ^[2].
• Check for sensor-interfering substances (Table 3) before attributing a recurrent high reading to hyperglycemia; agree when to use capillary backup readings ^[2].
• NICE reminds CGM users that capillary backup is needed to check sensor accuracy, to use as a backup when levels change quickly or the device fails, and to calibrate when required ^[3].

• Capillary backup: every CGM user needs enough test strips and to know when to use them; supplies should be reviewed at each visit ^[2][3].
• Frequency of BGM should be re-evaluated at every routine visit; monitoring alone does not lower glucose unless results are acted on ^[2].
• In type 1 diabetes NICE schedules 4 to 10 (sometimes more) meter checks daily; in low-resource or noninsulin settings fasting checks with planned dose changes are more realistic ^[3].
• AID and pump therapy need periodic review of manual-mode settings, sensor accuracy and lifestyle modes; CSII or AID access should be maintained across insurers and providers regardless of age or A1C ^[2].
• NICE advises that CGM should be delivered by teams with expertise, with attention to inequalities in access and uptake ^[3].

• Hypoglycemia from over-treatment guided by bad data: an inaccurate CGM or a faulty meter can cause unnecessary insulin or missed treatment. Confirm unexpected or symptomatic readings.
• Skin injury, contact dermatitis and infection at sensor or infusion sites.
• Infusion-set failure leading to ketosis or diabetic ketoacidosis.
• Lipohypertrophy and errant absorption from repeated sets.
• Alarm fatigue if CGM alerts are not configured sensibly; education at initiation is the countermeasure ^[2].
• Medicolegal and safety consequences of inaccurate readings in critical tasks: drivers, pilots, machinery operators and hospital inpatients need laboratory confirmation when in doubt.

CGM benefits in pregnancy include time in range and A1C goals in type 1 diabetes, and device indications have expanded for Dexcom G7, FreeStyle Libre 2 and FreeStyle Libre 3. Detailed glycemic goals, time-in-range targets in pregnancy and use of AID in pregnancy need the diabetes-in-pregnancy guidance; the Standards of Care point to section 15 ^[2]. NICE pregnancy targets for plasma glucose are outside the scope of this page and require the pregnancy-specific guideline ^[3].

CGM is recommended from onset and thereafter for children and adolescents on insulin or on therapy that can cause hypoglycemia, and insulin pumps are approved and usable in children. School staff need written plans, training and a backup for device failure. Early CGM initiation after diagnosis of type 1 diabetes reduces A1C and improves satisfaction ^[1][2].

Less stringent goals, with time in range above 50 percent and time below 70 mg/dL under 1 percent, are appropriate for complex or intermediate health states, because hypoglycemia fear, falls and cognitive impairment dominate in this group ^[1]. Hypoglycemia risk is highest in the oldest adults on insulin or sulfonylureas.

When erythrocyte turnover is disturbed, or in hemoglobin variants that make A1C invalid, monitor with CGM, blood glucose profiles or fructosamine. In sickle cell disease and homozygous hemoglobin variants A1C cannot be measured ^[1][3].

• A single blood glucose meter with enough strips can transform care if results are turned into dose, diet or activity changes.
• Strips should be bought unexpired from reliable sources; secondhand strips are unsafe ^[2].
• Needles of 4 to 5 mm length are efficient and lower intramuscular injection risk; syringes may be reused by the same person in resource-limited settings with proper cleansing and storage ^[2].
• Pumps and AID require training, repair capacity and reliable access to consumables; without these, MDI with a good meter and structured education remains the safer default.
• NICE reminds clinicians to monitor and improve inequalities in CGM access, which applies equally across health systems ^[3].
• Suspect altered A1C in anemia or hemoglobinopathy and switch to CGM or blood glucose profiles.

• Diabetes in pregnancy
• Chronic complications of diabetes
• Cardiac risk and diabetes (related vascular topics)
• Insulin pharmacology and dosing
• Hypoglycemia recognition and treatment
• Diabetic kidney disease
• Diabetes in children and adolescents
• Endocrine physical examination and laboratory testing

• Monitoring combines retrospective control (HbA1c), point-of-care steering (blood glucose) and continuous pattern information (CGM, with time in range).
• HbA1c reflects 2 to 3 months of average glucose; it is unreliable or unmeasurable when hemoglobin or red-cell turnover is disturbed, and alternatives must be used.
• For many nonpregnant adults the HbA1c goal is under 7 percent (under 53 mmol/mol), with time in range above 70 percent; more stringent or more lenient goals depend on health, hypoglycemia risk, age and life expectancy.
• CGM is recommended from onset in insulin-treated diabetes and during noninsulin treatment that risks hypoglycemia; used as close to daily as possible, uninterrupted.
• A valid CGM assessment needs 10 to 14 days of wear with 70 percent or more time active; standard goals include TIR above 70 percent, TBR under 4 percent for level 1 and under 1 percent for level 2, TAR under 25 percent for level 1, and CV 36 percent or lower in most adults.
• Everyone on CGM still needs an accurate meter; meters require approved devices, unexpired strips and awareness of interfering substances.
• Automated insulin delivery is the preferred insulin delivery method in type 1 diabetes and is suitable from diagnosis in adults, adolescents and children.
• Pump or AID complications include infusion-set failure with risk of ketoacidosis, skin injury, lipohypertrophy and dermatitis.
• Hypoglycemia prevention requires annual hypoglycemia-awareness screening, structured education and the availability of glucagon.
• Technology access and education determine outcomes as much as the device; inequalities in access should be audited and reduced.