The Diabetes Inbox: A Triage Guide for the 10 Most Common Patient Messages

A Cheat Sheet for Managing Remote Care Without a Full Visit

Dr Neeraj Manikath , claude.ai

Abstract

The management of diabetes has increasingly shifted toward remote monitoring and asynchronous communication, creating an unintended burden on healthcare providers through inbox management. Studies suggest that physicians spend up to 2 hours daily managing electronic messages, with diabetes-related queries comprising a significant proportion. This review provides evidence-based protocols for addressing the 10 most common diabetes-related patient messages, enabling efficient triage and management without necessitating a full clinic visit. By systematizing responses to recurring scenarios, clinicians can improve patient safety, glycemic outcomes, and their own professional sustainability.

Introduction

The digital transformation of healthcare has revolutionized diabetes management but has simultaneously created what many clinicians call "the second shift"—the hours spent after clinic managing patient portals, secure messages, and phone triage. Unlike scheduled visits, these interactions are unpredictable, uncompensated, and often lack clear protocols. For postgraduate physicians entering practice, mastering efficient inbox management is as critical as interpreting hemoglobin A1c values.

This article distills evidence-based approaches to the most common diabetes inbox scenarios into actionable protocols. The goal is not to replace clinical judgment but to provide a structured framework that reduces decision fatigue while maintaining patient safety and satisfaction.

1. "My Sugars Are Running High for 2 Days": The Sick-Day Rule Checklist

Clinical Context

Hyperglycemia during acute illness represents one of the most common—and potentially dangerous—scenarios in diabetes management. Illness-induced stress hormones (cortisol, catecholamines, glucagon, growth hormone) create insulin resistance, while decreased oral intake and medication adherence compound the problem. Studies show that up to 30% of diabetic ketoacidosis (DKA) cases are precipitated by concurrent illness.

The Triage Protocol

Immediate Red Flags (Require Same-Day Assessment or ED Referral):

Blood glucose persistently >400 mg/dL (22.2 mmol/L)
Presence of ketones (blood β-hydroxybutyrate >1.5 mmol/L or moderate/large urine ketones)
Signs of dehydration (decreased urine output, orthostatic symptoms)
Altered mental status or inability to maintain hydration
Severe abdominal pain, persistent vomiting
Kussmaul respirations or fruity breath odor

The Sick-Day Rule Checklist (For Stable Patients):

Never Stop Insulin: This is the cardinal rule. Even if not eating, basal insulin requirements persist—often increased during illness.
Hydration Protocol:
- Target 8 oz (240 mL) of sugar-free fluids hourly while awake
- If unable to eat solid food, switch to carbohydrate-containing fluids (juice, regular soda, broth) to prevent hypoglycemia
- Recommendation: Alternate between electrolyte solution and water
Glucose Monitoring Intensification:
- Check blood glucose every 2-4 hours (including overnight if severely elevated)
- Check ketones if glucose >250 mg/dL (13.9 mmol/L) or with nausea/vomiting
Insulin Adjustment Algorithm:
- For Type 1 diabetes: Continue basal insulin at usual dose. Add 10-20% supplemental rapid-acting insulin every 2-4 hours if glucose >250 mg/dL
- For Type 2 diabetes on insulin: Increase basal by 20% temporarily; add correction doses as needed
- For Type 2 diabetes on oral agents: Continue metformin if tolerating fluids; hold SGLT2 inhibitors due to DKA risk; consider adding temporary basal insulin if glucose consistently >300 mg/dL
Medication Considerations:
- Hold SGLT2 inhibitors (increased euglycemic DKA risk during illness)
- Continue metformin if no GI symptoms and adequate hydration
- Maintain GLP-1 agonists unless severe nausea/vomiting

Pearl: The "Rule of 1800"

For rapid-acting insulin sensitivity factor calculation: 1800 ÷ total daily insulin dose = mg/dL drop per 1 unit. Example: Patient on 60 units/day → 1800 ÷ 60 = 30 mg/dL drop per unit. This helps determine supplemental correction doses during sick days.

Follow-Up Protocol

Recheck in 24 hours via secure message
If no improvement in 48 hours or any red flags develop, schedule urgent visit

2. "I Have Low Sugars Every Afternoon": How to Identify and Correct the Pattern

Clinical Context

Recurrent hypoglycemia at predictable times suggests a medication-meal-activity mismatch rather than random variation. Pattern recognition is essential for appropriate intervention. Studies demonstrate that addressing predictable hypoglycemia reduces fear of hypoglycemia and improves overall glycemic control.

The Diagnostic Approach

Step 1: Data Collection via Secure Messaging Request structured information:

Timing of lows (exact times for past 7 days)
Glucose values (preferably with CGM trend data)
Timing of meals, snacks, and carbohydrate content
Medication administration times and doses
Physical activity patterns
Alcohol consumption (often forgotten factor)

Step 2: Pattern Classification

Type A: Postprandial Hypoglycemia (2-4 hours after meals)

Cause: Excessive prandial insulin or sulfonylurea for meal consumed
Solution: Reduce mealtime insulin by 10-20% or adjust insulin-to-carb ratio; consider switching sulfonylurea to DPP-4 inhibitor or GLP-1 agonist

Type B: Pre-Meal Hypoglycemia

Cause: Excessive basal insulin or insulin stacking from previous boluses
Solution: Reduce basal insulin by 10-20%; ensure 4-5 hour spacing between rapid-acting doses; add structured snacks

Type C: Exercise-Associated Hypoglycemia

Cause: Insulin action enhanced by increased muscle glucose uptake
Solution: Reduce prandial insulin by 25-50% if exercise within 2 hours post-meal; consume 15-30g carbs before prolonged activity; consider temporary basal rate reduction with pumps

Type D: Nocturnal/Fasting Hypoglycemia with Afternoon Rebound

Cause: Counter-regulatory hormone response (cortisol, growth hormone) causing morning hyperglycemia, leading to overcorrection and afternoon hypoglycemia
Solution: Address overnight basal rates; avoid aggressive morning corrections; use CGM to document pattern

The Correction Protocol

For Insulin Users:

Adjust the insulin dose acting during the hypoglycemic period (4-6 hours back from event for rapid-acting; 24 hours review for basal)
Apply the "10-20% rule": Initial adjustments should be 10-20% of current dose
Change one variable at a time and assess for 3 days

For Sulfonylurea/Meglitinide Users:

Consider splitting doses or switching to agents with lower hypoglycemia risk (DPP-4 inhibitors, GLP-1 agonists, SGLT2 inhibitors)
Evaluate for renal dysfunction prolonging drug half-life

Oyster: The "Dawn Phenomenon Overcorrection Cycle"

Many afternoon lows stem from aggressive correction of dawn phenomenon (physiologic morning hyperglycemia). Patients see 180 mg/dL at 7 AM, aggressively correct, then crash at 2 PM. Solution: Accept morning glucose 140-180 mg/dL without correction if pattern of afternoon lows exists. Adjust evening basal instead.

Hack: The "Snack-First Strategy"

For patients with predictable afternoon lows who resist medication changes: Prescribe a mandatory 15g carb snack 1 hour before typical low (e.g., 2 PM snack if lows at 3 PM). This often breaks the cycle while medication adjustments are being optimized.

3. "I Forgot My Insulin": Emergency Dosing and Pharmacist Scripting

Clinical Context

Medication access issues create urgent safety risks. The average patient with Type 1 diabetes will develop DKA within 24-48 hours without insulin. Providing immediate guidance while arranging prescription access is critical.

The Triage Assessment

Critical Questions:

Type of diabetes (Type 1 vs. Type 2)?
Which insulin was forgotten (basal, prandial, or both)?
Last insulin dose and timing?
Current blood glucose and presence of ketones?
Access to pharmacy (hours until can obtain)?
Insurance/financial barriers?

Immediate Management by Diabetes Type

Type 1 Diabetes (URGENT):

This is a medical emergency—risk of DKA within 24 hours
Options in order of preference:
1. Emergency pharmacy bridge supply: Call patient's pharmacy directly; provide verbal authorization for 3-day emergency supply; most states allow this without new prescription
2. Alternative insulin source: If patient has "wrong" insulin at home (e.g., has prandial but needs basal), provide temporary conversion guidance
3. ED referral: If no access within 4-6 hours and glucose >300 mg/dL or ketones present

Type 2 Diabetes:

Assess whether insulin-dependent (C-peptide negative, long diabetes duration, lean) vs. insulin-supplemented
For insulin-supplemented Type 2: Temporarily increase oral agents and monitor closely—most can safely miss 24-48 hours of insulin
For insulin-dependent Type 2: Treat as Type 1

Pharmacist Scripting Template

Provide this language to patients for pharmacy communication:

"My doctor authorized an emergency 3-day supply of [insulin name and strength] while my prescription is being processed. Please contact Dr. [Your Name] at [phone number] if verification is needed. My blood glucose is currently [value] and I last took insulin [timing]."

Temporary Insulin Substitution Protocols

If Patient Has Rapid-Acting But Needs Basal:

Not ideal, but can use rapid-acting insulin every 6 hours to provide continuous coverage
Dose: Calculate usual total daily basal dose ÷ 4 doses
Example: If on glargine 30 units daily → give aspart 7-8 units every 6 hours
Requires glucose checks every 4 hours and snack with each dose

If Patient Has Basal But Needs Rapid-Acting:

Increase basal by 20-30% temporarily
Use sliding scale for corrections only (not for meal coverage)
Advise lower-carbohydrate meals until prandial insulin obtained

If Patient Has No Insulin:

Over-the-counter options: NPH and regular insulin (Novolin N, Novolin R) available without prescription at most pharmacies ($25-50/vial)
NPH dosing: Start 0.3 units/kg/day divided as 2/3 morning, 1/3 evening
Regular insulin: Use for meal coverage, give 30 minutes before eating (older analog, requires planning)

Pearl: The "Expired Insulin Rescue"

Insulin maintains 90% potency for 2-3 months past expiration date if refrigerated. In true emergencies, expired insulin is safer than no insulin. Instruct patient to increase doses by 10-20% to compensate for reduced potency.

Insurance Navigation Script

For patients with insurance barriers:

Contact pharmacy benefits: "Request vacation override or lost medication replacement"
Manufacturer assistance: All insulin manufacturers have patient assistance programs (turnaround 24-72 hours for emergencies)
Samples: If you have office samples, arrange pickup

Follow-Up

Verify insulin obtained within 6 hours for Type 1 diabetes
Address root cause: Set up automatic refills, extra vial at work/relatives, consider insulin pump with reservoir backup

4. "My CGM Is Inaccurate": A Step-by-Step Troubleshooting Guide

Clinical Context

Continuous glucose monitors (CGMs) have revolutionized diabetes management, but sensor errors create anxiety and potentially dangerous decisions. Understanding the 15% accuracy variation inherent in CGM technology helps set appropriate expectations. The FDA allows ±15% variation from laboratory glucose values, meaning a true glucose of 200 mg/dL could read 170-230 mg/dL and still be "accurate."

Triage Questions to Ask

How do you know it's inaccurate? (Comparison to fingerstick? Symptoms?)
How long has the sensor been in place?
What is the CGM reading vs. fingerstick meter reading?
Is glucose stable, rising, or falling?
Any recent medications (acetaminophen, vitamin C, aspirin)?
Sensor insertion site issues (bleeding, redness, wetness)?

The Troubleshooting Algorithm

Step 1: Verify True Inaccuracy

CGM lags behind blood glucose by 5-15 minutes (interstitial vs. capillary glucose)
During rapid glucose changes, discrepancies up to 50 mg/dL are expected
Rule: Only compare CGM to fingerstick when glucose is stable (flat arrow) for 15+ minutes
If glucose rising/falling: Wait 15 minutes after fingerstick, then reassess

Step 2: Assess Magnitude of Discrepancy

Minor Discrepancy (<20% difference):

Example: CGM 180 mg/dL, fingerstick 210 mg/dL
Action: No intervention needed; this is within expected accuracy range
Patient education: CGM is a trend monitor, not a replacement for all fingersticks

Moderate Discrepancy (20-30% difference):

Example: CGM 150 mg/dL, fingerstick 210 mg/dL
Action: Perform CGM calibration if system allows (Dexcom G6/G7 self-calibrate; Guardian 3 requires manual calibration)
If persistent after calibration, consider sensor replacement

Severe Discrepancy (>30% difference or opposite direction):

Example: CGM 80 mg/dL, fingerstick 200 mg/dL
Action: Sensor failure likely—replace sensor and use fingersticks for insulin dosing decisions
Document lot number for manufacturer complaint

Step 3: Identify Common Causes

Sensor-Related:

Compression low: False low readings when sleeping on sensor (appears as sudden drop followed by recovery). Solution: Change sensor site to area less prone to compression (abdomen over arm)
End-of-life drift: Sensors lose accuracy in final 24 hours. Dexcom G6 lasts 10 days; check timing
Insertion trauma: Bleeding during insertion damages sensor. If blood visible under adhesive, replace sensor immediately
Adhesive failure: Water/sweat ingress ruins sensor. Use over-patches (Skin Grip, ExpressionMed) in high-humidity environments

Patient Technique:

Poor fingerstick technique: Dirty fingers (especially glucose-containing foods/drinks), insufficient blood, squeezing finger. Always wash hands first
Older meter strips: Expired or humidity-damaged strips cause meter inaccuracy
Different blood source: Venous vs. capillary blood glucose can differ by 10%; always use fingertips, not forearm

Medication Interference:

Acetaminophen: Causes falsely elevated readings in older CGM systems (Dexcom G4/G5); newer systems less affected. If taking >1g acetaminophen, verify with fingerstick
Vitamin C >500mg/day: Can affect some electrochemical sensors
Hydroxyurea: May interfere with certain CGM enzymes
Mannitol/xylose: Used in some medications, can cross-react

Physiological:

Dehydration: Concentrated interstitial fluid can cause discrepancies
Edema: Fluid retention dilutes interstitial glucose
Rapidly changing glucose: CGM always lags—trust symptoms over device during rapid changes

The Action Protocol

For Symptomatic Hypoglycemia with Normal CGM:

Always treat symptoms regardless of CGM reading
Rule: "Treat the patient, not the number"
Give 15g fast-acting carbs, recheck with fingerstick in 15 minutes

For Asymptomatic CGM Alert with Normal Fingerstick:

If stable: Trust fingerstick, monitor CGM trend
If discrepancy persists >2 readings: Replace sensor

For Hyperglycemia Discrepancy:

Less urgent than hypoglycemia
Use fingerstick for insulin dosing decisions
Replace sensor if pattern continues >4 hours

Hack: The "Fingerstick Calibration Schedule"

For patients frequently reporting inaccuracy, establish a structured calibration schedule:

Before meals and bedtime (matching insulin dosing times)
Only calibrate when glucose stable (flat arrow)
Never calibrate during first 24 hours after insertion or when glucose rapidly changing
This reduces anxiety and provides actionable data

Manufacturer Contact Indications

Sensor fails within first 3 days (usually replaced free)
Consistent >30% inaccuracy despite troubleshooting
Adhesive failure before end of sensor life
Persistent "no reading" errors

Pearl: The "Warm-Up Period Reality"

Most CGMs require 2-24 hours "warm-up" after insertion for optimal accuracy. During this time, 30-40% discrepancy is common. Counsel patients: "New sensors are like new shoes—they need breaking in. Day 1 readings may be off; by day 2-3, they'll stabilize."

5. "Can I Take This New Medication?": The Drug Interaction Red Flags

Clinical Context

Polypharmacy is the norm in diabetes management, with average patients taking 7-10 medications. New prescriptions from other specialists or over-the-counter additions create drug interaction risks. Systematic screening prevents adverse events.

The Triage Framework

Essential Information to Obtain:

Exact medication name, dose, and frequency
Prescribing indication
Prescribing physician (specialist?)
Current complete medication list with diabetes regimen
Timing: When would medication be taken relative to diabetes medications?

High-Priority Drug Interactions in Diabetes

Category 1: Medications That Directly Alter Glucose

Hyperglycemia-Inducing Agents:

Glucocorticoids (prednisone, dexamethasone): Dose-dependent hyperglycemia, primarily affecting post-prandial glucose.
- Management: Increase prandial insulin 50-100% or add NPH to cover daytime hours; adjust based on steroid timing (morning dosing affects day more than night)
- Pearl: Inhaled/topical steroids rarely affect glucose; focus on systemic formulations
Atypical antipsychotics (olanzapine, quetiapine, risperidone): Increase insulin resistance and appetite
- Management: Increase basal insulin 20-30%; monitor for metabolic syndrome; consider metformin addition if not already prescribed
Thiazide diuretics: Mild hyperglycemic effect at higher doses
- Generally safe to continue; monitor glucose for 2 weeks after initiation
Beta-blockers (particularly non-selective): Mask hypoglycemia symptoms, impair glycogenolysis
- Management: Intensify glucose monitoring; educate on autonomic-independent symptoms (confusion, difficulty concentrating)

Hypoglycemia-Potentiating Agents:

Fluoroquinolones (especially gatifloxacin, but all quinolones): Bidirectional glucose effects
- Action: Increase glucose monitoring to 4 times daily; reduce sulfonylurea dose by 50% preventively
Pentamidine: Causes severe hypoglycemia followed by hyperglycemia (pancreatic beta-cell toxicity)
- Requires hospitalization for administration in diabetes patients
Alcohol: Impairs hepatic gluconeogenesis
- Education: Never drink on empty stomach; consume with carb-containing meal; increased hypo risk 8-12 hours post-consumption

Category 2: Medication Absorption and Efficacy Issues

SGLT2 Inhibitors + Diuretics:

Combined dehydration risk, especially with loop diuretics
Action: Counsel on hydration (80 oz/day); hold during illness; monitor kidney function

GLP-1 Agonists + Other Medications:

Delayed gastric emptying affects absorption of oral medications
Particularly important: Levothyroxine, oral contraceptives, antibiotics requiring specific timing
Solution: Take time-sensitive medications 1 hour before GLP-1 injection or 4 hours after

Metformin + Contrast Dye:

Lactic acidosis risk if acute kidney injury occurs post-contrast
Protocol: Hold metformin day of and 48 hours after contrast; verify creatinine normal before resuming

Category 3: Renal Function Considerations

Many diabetes medications are renally cleared. New nephrotoxic medications require reassessment:

NSAIDs:

Reduce eGFR, particularly with ACE inhibitors/ARBs
Action: If chronic NSAID needed, recheck creatinine in 1 week; dose-adjust diabetes medications if eGFR changes

Aminoglycosides, vancomycin:

Monitor renal function closely
Dose-reduce sulfonylureas, SGLT2 inhibitors if eGFR drops

Category 4: Cardiovascular Drug Interactions

Statins:

No direct interaction; generally beneficial
Exception: High-dose statins may worsen glucose control in ~10% of patients
Action: Monitor A1c; modest increase (0.2-0.3%) acceptable given CV benefit

Fibrates + Sulfonylureas:

Fibrates displace sulfonylureas from protein binding, increasing free (active) drug
Action: Reduce sulfonylurea dose by 50% when initiating fibrate

Category 5: Over-the-Counter and Supplements

Niacin (>1g/day):

Significant hyperglycemia
Action: Discourage use or increase diabetes medications 20-30%

Chromium, cinnamon, alpha-lipoic acid:

Marketed for "glucose control"—minimal evidence
Generally safe but may potentiate insulin/sulfonylureas modestly
Action: Counsel that these are not substitutes for proven therapies; monitor for hypoglycemia if patient insists on use

The Response Template for Common Scenarios

Scenario: Patient prescribed prednisone 40mg daily for 5 days by urgent care

"Prednisone will significantly raise your blood sugars, especially 4-8 hours after taking it. Here's your adjusted plan:

Double your mealtime insulin doses for breakfast and lunch
Check your glucose 4 times daily (before meals and bedtime)
If glucose goes above 300 mg/dL, add 10% extra basal insulin
These changes apply only while taking prednisone and for 2 days after finishing
If sugars remain above 250 mg/dL for more than 12 hours despite adjustments, contact me
After finishing the steroid, return to your normal insulin doses"

Scenario: Cardiologist started amiodarone

"Amiodarone doesn't directly interact with your diabetes medications, but monitor for new thyroid issues—this drug can affect thyroid function, which impacts glucose control. I'll add a TSH check to your next lab order. Continue all diabetes medications as prescribed. The bigger concern is coordinating with your cardiologist about your current medication list—I'll send them a note."

Hack: The "Specialist Loop-Closing Script"

When another provider prescribes medication requiring diabetes regimen adjustment, send a secure message:

"Dr. [Specialist] prescribed [medication]. I've adjusted your diabetes management as follows: [specific changes]. Please ensure all your doctors have your current medication list. I'm sending Dr. [Specialist] a summary of the changes. Continue seeing them as scheduled."

This prevents patients from feeling caught between providers and documents coordination.

Oyster: The "Supplement Disclosure Problem"

Many patients don't consider supplements "real medications." Specifically ask: "Are you taking any vitamins, herbs, minerals, teas, or products from health food stores?" Common diabetes-relevant supplements:

Berberine (actually lowers glucose—may cause hypoglycemia)
Bitter melon
Fenugreek
St. John's Wort (affects liver metabolism of multiple drugs)

Red Flag Medications Requiring Immediate Discussion

Call patient (don't just message) if prescribed:

Fluoroquinolones in sulfonylurea users
High-dose steroids (>20mg prednisone equivalent) in Type 1 diabetes
Nephrotoxic drugs in patients with eGFR <45
Any medication labeled "may affect glucose" in patients with history of severe hypoglycemia or hypoglycemia unawareness

Clinical Decision Support Tools

Recommend patients use:

Drugs.com interaction checker (patient-friendly)
Medscape drug interaction checker (more comprehensive)

For your own reference:

Lexicomp drug interactions database
UpToDate drug interactions tool
FDA package inserts for detailed pharmacokinetics

Conclusion: Building Inbox Sustainability

Efficient diabetes inbox management requires balancing accessibility with sustainability. The protocols outlined provide structure for the most common scenarios, but each interaction should be documented as a brief encounter in the medical record with appropriate coding for time spent (CPT codes 99421-99423 for online digital E/M services are increasingly reimbursed).

Systems-Level Solutions

To reduce inbox burden long-term:

Develop templated responses (SmartPhrases/Dot Phrases) for common scenarios
Empower nursing staff with standing orders for routine scenarios
Implement patient self-management pathways using algorithms
Schedule dedicated "inbox time" rather than reactive management throughout the day
Use patient portals for structured data collection before responding

The 80/20 Rule in Practice

These five scenarios represent approximately 80% of diabetes inbox messages. Mastering efficient, safe responses enables better care delivery while protecting physician well-being. The remaining 20% of complex scenarios will always require clinical judgment—but even those decisions become faster when routine queries are handled systematically.

Key Takeaways for Postgraduate Physicians

Sick days are always urgent in Type 1 diabetes—develop same-day assessment pathways
Recurrent hypoglycemia patterns are fixable—systematic data collection prevents guesswork
Insulin access interruptions require immediate action—know your emergency prescribing protocols
CGM accuracy concerns are usually explainable—structured troubleshooting reduces anxiety
Drug interactions in diabetes are predictable—focus on glucose-altering agents, GI effects, and renal function

Final Pearl: The "Inbox Efficiency Mindset"

Not every message requires a perfect solution—many require rapid triage and a follow-up plan. Aim for "safe and good enough today" rather than "perfect resolution immediately." Your inbox is sustainable when you protect time for complex patients who truly need deep engagement.

References

Holman N, et al. Current prevalence of Type 1 and Type 2 diabetes in adults and children in the UK. Diabet Med. 2015;32(9):1119-1120.
Kitabchi AE, et al. Hyperglycemic crises in adult patients with diabetes. Diabetes Care. 2009;32(7):1335-1343.
Danne T, et al. International consensus on use of continuous glucose monitoring. Diabetes Care. 2017;40(12):1631-1640.
American Diabetes Association. Standards of Medical Care in Diabetes—2024. Diabetes Care. 2024;47(Suppl 1):S1-S321.
Seaquist ER, et al. Hypoglycemia and diabetes: a report of a workgroup of the American Diabetes Association and The Endocrine Society. Diabetes Care. 2013;36(5):1384-1395.
Pfützner A, et al. Technical aspects of the Parkes error grid. J Diabetes Sci Technol. 2013;7(5):1275-1281.
Luna B, Feinglos MN. Drug-induced hyperglycemia. JAMA. 2001;286(16):1945-1948.
Parekh TM, et al. Hypoglycemia after antimicrobial drug prescription for older patients using sulfonylureas. JAMA Intern Med. 2014;174(10):1605-1612.
Rodbard HW, et al. Statement by an American Association of Clinical Endocrinologists/American College of Endocrinology consensus panel on type 2 diabetes mellitus: an algorithm for glycemic control. Endocr Pract. 2009;15(6):540-559.
Tai MM, et al. Reducing the frequency of hypoglycemia in patients with diabetes using an educational programme. Diabetes Metab Syndr Obes. 2015;8:423-432.
Umpierrez GE, Korytkowski M. Diabetic emergencies—ketoacidosis, hyperglycaemic hyperosmolar state and hypoglycaemia. Nat Rev Endocrinol. 2016;12(4):222-232.
Battelino T, et al. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time in range. Diabetes Care. 2019;42(8):1593-1603.
Davies MJ, et al. Management of hyperglycemia in type 2 diabetes: a consensus report by the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2018;41(12):2669-2701.
Hirsch IB, Bode BW. Therapeutic implications of continuous glucose monitoring. Diabetes Technol Ther. 2017;19(S2):S-3-S-13.
Tamez HE, et al. Drug-induced diabetes mellitus. Expert Opin Drug Saf. 2016;15(10):1401-1414.