In an adult without renal or hepatic impairment, not pregnant, and without frequent hypoglycemia, how much should insulin be increased at each titration?

How Much Should You Increase Insulin at a Time?

For adults without renal or hepatic impairment, not pregnant, and without frequent hypoglycemia, increase basal insulin by 2–4 units every 3 days based on fasting glucose levels, and increase prandial insulin by 1–2 units (or 10–15%) every 3 days based on postprandial glucose readings. 1

Basal Insulin Titration Algorithm

The most widely validated approach uses a structured, glucose-driven protocol:

• If fasting glucose is 140–179 mg/dL: Increase basal insulin by 2 units every 3 days 1, 2
• If fasting glucose is ≥180 mg/dL: Increase basal insulin by 4 units every 3 days 1, 2
• Target fasting glucose: 80–130 mg/dL 1, 2

An alternative titration method recommended by some guidelines is to increase by 10–15% of the current dose once or twice weekly until fasting blood glucose targets are met 1. This percentage-based approach may be particularly useful at higher insulin doses where fixed-unit increases become proportionally smaller.

Prandial (Mealtime) Insulin Titration

When adding or adjusting rapid-acting insulin:

• Increase each meal dose by 1–2 units every 3 days based on 2-hour postprandial glucose readings 1, 2
• Alternatively, increase by 10–15% of the current dose every 3 days 1
• Target postprandial glucose: <180 mg/dL 1, 2

Critical Threshold: When to Stop Escalating Basal Insulin

Do not continue increasing basal insulin beyond 0.5–1.0 units/kg/day without addressing postprandial hyperglycemia. 1, 3 This threshold is crucial because:

• When basal insulin approaches 0.5 units/kg/day (approximately 35–50 units for most adults), further increases often produce diminishing returns 1, 2
• Continuing to escalate basal insulin beyond this point leads to "overbasalization"—a dangerous pattern characterized by increased hypoglycemia risk without improved glycemic control 1, 3
• At this threshold, add prandial insulin (starting with 4 units before the largest meal) rather than further increasing basal insulin 1, 2

Clinical signs of overbasalization include: basal dose >0.5 units/kg/day, bedtime-to-morning glucose differential ≥50 mg/dL, hypoglycemia episodes, and high glucose variability 1.

Hypoglycemia Response Protocol

If any unexplained hypoglycemia (<70 mg/dL) occurs, immediately reduce the implicated insulin dose by 10–20%. 1, 2 This reduction should be made before the next scheduled dose, not delayed until the next clinic visit.

If more than two fasting glucose values per week are <80 mg/dL, decrease the basal insulin dose by 2 units 1.

Starting Doses for Context

While your question focuses on titration increments, understanding starting doses provides important context:

• Type 2 diabetes (insulin-naive): Start with 10 units once daily or 0.1–0.2 units/kg/day 1, 4
• Severe hyperglycemia (HbA1c ≥9% or glucose ≥300 mg/dL): Start with 0.3–0.5 units/kg/day total daily dose, split 50% basal and 50% prandial 1
• Type 1 diabetes: Typically 0.5 units/kg/day total, with approximately 40–50% as basal and 50–60% as prandial 1

Monitoring Requirements

• Check fasting glucose daily during active titration to guide basal insulin adjustments 1, 2
• Check 2-hour postprandial glucose after meals to guide prandial insulin adjustments 1, 2
• Reassess every 3 days during active titration 1
• Reassess HbA1c every 3 months once stable 1

Common Pitfalls to Avoid

• Never use sliding-scale insulin as monotherapy—it is condemned by major diabetes guidelines as ineffective and dangerous 1
• Do not delay titration—timely dose adjustment is essential for achieving glycemic goals 1
• Do not make adjustments more frequently than every 3 days for standard basal insulins, as this doesn't allow sufficient time to assess the full effect 1
• Never give rapid-acting insulin at bedtime as a sole correction dose—this markedly increases nocturnal hypoglycemia risk 1

Evidence Quality Note

The 2–4 unit titration algorithm for basal insulin is derived from the landmark Treat-to-Target study and has been validated across multiple randomized controlled trials involving thousands of patients 5. The AT.LANTUS study demonstrated that both physician-managed and patient-managed titration using these increments successfully achieve HbA1c <7% with acceptable hypoglycemia rates 5.