Sulfonylureas Monitoring Bundle: The Complete Prescriber Reference

Medical lab testing image for Sulfonylureas Monitoring Bundle: The Complete Prescriber Reference

At a glance

  • Drug class / Second-generation sulfonylureas (glipizide, glyburide, glimepiride) for type 2 diabetes
  • Prototype agent / Glipizide (Glucotrol), shortest active half-life in class
  • HbA1c reduction / 1.0 to 2.0 percentage points below baseline when added to metformin
  • Primary safety concern / Hypoglycemia, especially with glyburide in older adults and CKD
  • Mechanism / ATP-sensitive K+ channel closure on beta cells, forcing insulin release
  • Renal threshold / Avoid glyburide if eGFR <60 mL/min/1.73 m²; titrate glipizide cautiously at eGFR <30
  • Cardiovascular signal / UKPDS 33 showed metformin superior to sulfonylurea monotherapy for CV outcomes in obese patients
  • Monitoring frequency / HbA1c every 3 months until at goal, then every 6 months; BMP every 6 to 12 months
  • Key drug interaction / Fluconazole doubles glipizide AUC via CYP2C9 inhibition
  • Weight effect / Average 2 to 5 kg weight gain over the first year of therapy

What Is the Sulfonylureas Drug Class?

Sulfonylureas are insulin secretagogues that close ATP-sensitive potassium (K-ATP) channels on pancreatic beta cells, depolarizing the membrane and triggering calcium-mediated insulin exocytosis. The result is glucose-independent insulin release, which explains their hypoglycemia liability even in the fasting state. First approved in the 1950s, they remain among the most widely prescribed oral agents for type 2 diabetes globally [1].

Generation Differences

First-generation agents (chlorpropamide, tolbutamide) carry longer half-lives and higher rates of drug interactions. Second-generation agents, including glipizide, glyburide, and glimepiride, are preferred today because of greater potency at lower doses and better-characterized pharmacokinetics [2].

Mechanism Details That Drive Monitoring Decisions

Glyburide produces active metabolites that accumulate in renal impairment, extending its hypoglycemic effect unpredictably. Glipizide is metabolized by CYP2C9 to inactive compounds, making it safer in mild-to-moderate CKD. Glimepiride shares the inactive-metabolite profile but still requires dose reduction below eGFR 30 [3]. These mechanistic differences are the foundation for every renal dosing rule in this bundle.

Place in Current Guidelines

The 2024 American Diabetes Association Standards of Care list sulfonylureas as acceptable add-on therapy after metformin when cost is a barrier to newer agents. The document states: "Sulfonylureas are effective glucose-lowering medications but are associated with hypoglycemia and weight gain, and agents with lower hypoglycemia risk are preferred when available" [4]. The ADA places GLP-1 receptor agonists and SGLT2 inhibitors above sulfonylureas for patients with established cardiovascular disease or high hypoglycemia risk, but acknowledges that price parity rarely exists in practice.


Baseline Assessment Before Starting a Sulfonylurea

Every patient needs a pre-prescription workup. Skipping any component leaves you without the reference values you will need when a problem arises at month three.

Required Labs at Baseline

Order these before writing the first prescription [5]:

  • HbA1c to establish starting glycemic burden and set a titration target
  • Comprehensive metabolic panel (CMP) for eGFR, creatinine, AST, ALT, and albumin
  • Complete blood count (CBC) because rare sulfonylurea-related hemolytic anemia requires a baseline
  • Fasting glucose to assess the immediate hypoglycemia risk window at initiation

An eGFR below 30 mL/min/1.73 m² is a contraindication for glyburide and requires a 50% dose reduction for glimepiride per FDA labeling [6]. Liver transaminases more than three times the upper limit of normal are a relative contraindication; hepatic impairment slows sulfonylurea metabolism and increases free drug exposure.

Patient History Checklist

Ask specifically about [4]:

  • Prior episodes of hypoglycemia and whether the patient recognized them
  • Meal timing variability (shift workers and elderly patients with inconsistent eating are at highest risk)
  • Concurrent use of fluconazole, trimethoprim-sulfamethoxazole, clarithromycin, or fluoroquinolones, all of which raise sulfonylurea plasma levels
  • Alcohol use, because ethanol blocks gluconeogenesis and magnifies hypoglycemia severity

Establishing a Hypoglycemia Baseline

Ask the patient to describe what "low blood sugar" feels like for them. Patients with autonomic neuropathy or a history of hypoglycemia unawareness need continuous glucose monitoring (CGM) from day one rather than a standard SMBG schedule [7].


The Core Monitoring Schedule

Structured follow-up converts a high-risk drug class into a manageable one. The table below shows the minimum monitoring cadence endorsed by ADA and AACE guidelines [4][8].

| Timepoint | Test | Action Threshold | |---|---|---| | Baseline | HbA1c, CMP, CBC, fasting glucose | Defer if eGFR <30 or LFTs >3x ULN | | 4 weeks | Fasting glucose, weight, SMBG log review | Up-titrate if fasting BG >130 mg/dL | | 3 months | HbA1c, eGFR | Step down dose if HbA1c <7.0% or hypoglycemia reported | | 6 months | HbA1c, CMP | Reassess class if eGFR has fallen below dosing threshold | | 12 months | HbA1c, CMP, CBC, weight | Annual safety review | | Every 12 months | Ophthalmology, foot exam | Per ADA microvascular standards |

HbA1c Targets and Over-Treatment Risk

The ADA target for most adults is HbA1c below 7.0%, but the 2024 Standards explicitly warn against targeting below 6.5% in older adults or patients with hypoglycemia unawareness, because sulfonylureas cannot modulate their insulin output in response to falling glucose [4]. The ACCORD trial (N=10,251) found that intensive glucose lowering to HbA1c below 6.0% with agents that included sulfonylureas increased all-cause mortality compared with standard care (hazard ratio 1.22, P<0.001) [9]. That signal hardened the case for individualized HbA1c targets, especially in patients over 70.

SMBG Frequency by Agent

Glyburide requires twice-daily fasting and pre-dinner glucose checks for the first 4 weeks. Glipizide's shorter half-life allows once-daily fasting checks after the titration period stabilizes. Patients on glimepiride once daily can use fasting-only SMBG once they have achieved 4 consecutive weeks below target without hypoglycemia [8].


Hypoglycemia: Recognition, Grading, and Response Protocols

Hypoglycemia is the rate-limiting adverse effect of this class. The Endocrine Society's 2023 clinical practice guideline on hypoglycemia in adults defines three levels [10]:

  • Level 1: Glucose 54 to 70 mg/dL, addressable with 15 g fast-acting carbohydrate
  • Level 2: Glucose below 54 mg/dL, requires immediate treatment and clinical review
  • Level 3: Severe cognitive impairment requiring third-party assistance, requires glucagon or IV dextrose

Who Is at Highest Risk?

Risk factors that compound sulfonylurea hypoglycemia [10][11]:

  • Age above 65 (impaired counter-regulatory response)
  • eGFR below 60 mL/min/1.73 m² (reduced renal gluconeogenesis, accumulation of active metabolites with glyburide)
  • Caloric restriction or gastroparesis
  • Beta-blocker use (masks adrenergic warning symptoms)
  • Alcohol consumption within 12 hours

The ORIGIN trial (N=12,537) found that sulfonylurea-containing regimens produced severe hypoglycemia in 5.7 events per 100 patient-years, compared with 1.7 events per 100 patient-years in insulin glargine arms after 6.2 years of follow-up [12]. That comparison is counterintuitive but reflects glyburide's unpredictable pharmacokinetics in the study population.

The 15-15 Rule and When It Fails

Standard treatment: 15 g of glucose (4 oz juice, 3 to 4 glucose tablets), recheck in 15 minutes [4]. If glucose remains below 70 mg/dL, repeat. Sulfonylurea-induced hypoglycemia, unlike short-acting insulin-induced hypoglycemia, can persist for 12 to 24 hours with glyburide because of ongoing insulin secretion from residual drug effect. Patients and caregivers must be told explicitly that resolution of symptoms does not mean the episode is over.

Glucagon Prescribing Alongside Sulfonylureas

Any patient on glyburide or glimepiride should have a glucagon kit or nasal glucagon (Baqsimi 3 mg) prescribed at the same visit [10]. Prescribe to the patient and give the administration instructions to a household member or caregiver. This is not optional for patients with prior Level 2 or Level 3 events.


Renal Dosing and CKD Monitoring

Kidney function determines which sulfonylurea is safe and at what dose. Check eGFR at every 6-month visit for patients with baseline CKD stage 3 or above [6].

Agent-Specific eGFR Thresholds

| Agent | eGFR ≥60 | eGFR 30 to 59 | eGFR 15 to 29 | eGFR <15 | |---|---|---|---|---| | Glipizide | Full dose | Use with caution, start low | 2.5 mg daily max | Avoid | | Glimepiride | Full dose | 1 mg daily, titrate slowly | 1 mg daily, no uptitration | Avoid | | Glyburide | Full dose | Avoid (active metabolites) | Avoid | Avoid |

FDA labeling for glimepiride specifies initiating at 1 mg/day in patients with renal impairment and up-titrating based on fasting glucose response only [6]. The 2023 KDIGO diabetes management guideline explicitly recommends against glyburide in any patient with CKD stage 3b or worse [13].

Acute Kidney Injury Protocols

If a patient on a sulfonylurea develops an acute illness with vomiting, fever, or diarrhea, instruct them to hold the drug until they are eating and drinking normally. Dehydration reduces renal clearance acutely, and continued dosing during illness causes disproportionate hypoglycemia. This "sick day rule" should be part of every patient education session [13].


Hepatic Impairment Considerations

The liver is the primary site of sulfonylurea metabolism and the primary site of counter-regulatory gluconeogenesis. Hepatic impairment impairs both, creating a double risk. Use with caution in Child-Pugh class A, and avoid in Child-Pugh B or C [5].

Monitoring Liver Function on Therapy

A CMP at baseline and at 6 months catches drug-related hepatotoxicity early. Sulfonylurea-associated cholestatic jaundice is rare but documented, with an estimated incidence of 1 in 10,000 patient-years for chlorpropamide and similar estimates for second-generation agents [14]. A transaminase rise above two times baseline on follow-up warrants a rechallenge discussion and consideration of switching to a DPP-4 inhibitor or basal insulin.


Cardiovascular Safety Evidence

UKPDS 33 and Its Legacy

The United Kingdom Prospective Diabetes Study 33 (UKPDS 33, N=3,867) remains the foundational cardiovascular trial for sulfonylureas. Intensive glucose control with glibenclamide or chlorpropamide reduced any diabetes-related endpoint by 12% (P<0.0001) compared to conventional diet therapy, but did not significantly reduce myocardial infarction as a standalone outcome [15]. The metformin subset showed superior CV benefit. That discordance shaped decades of prescribing guidelines.

Post-UKPDS Evidence

The TOSCA.IT trial (N=3,028) compared sulfonylurea (mainly glipizide or glimepiride) to pioglitazone as add-on to metformin over a median 57 months. Sulfonylureas and pioglitazone showed equivalent rates of the composite CV endpoint (hazard ratio 0.96, 95% CI 0.74 to 1.26), but the sulfonylurea arm had significantly more hypoglycemic events [16]. This provides reasonable reassurance that modern second-generation sulfonylureas carry no excess cardiovascular mortality signal versus an active comparator, while reinforcing that hypoglycemia mitigation remains the primary management goal.

QTc Monitoring Note

Glibenclamide (glyburide) has a small QTc-prolonging effect at high doses. Routine ECG is not required for all patients, but check a baseline ECG before initiating glyburide in patients already on QTc-prolonging agents (fluoroquinolones, antipsychotics, amiodarone) [17].


Drug Interactions Requiring Active Monitoring

CYP2C9 mediates the oxidative metabolism of all three second-generation sulfonylureas. Inhibitors raise plasma levels; inducers lower them [3][18].

High-Priority Inhibitors

  • Fluconazole: Increases glipizide AUC by approximately 100% in healthy volunteers [18]. Avoid co-prescription if possible; if unavoidable, halve the sulfonylurea dose and increase SMBG to twice daily during the antifungal course.
  • Trimethoprim-sulfamethoxazole: Inhibits CYP2C9 and competes for renal tubular secretion of glipizide. Risk of severe hypoglycemia within 48 hours of co-administration [3].
  • Clarithromycin: CYP3A4 and CYP2C9 inhibition; risk applies mainly to glimepiride.
  • Gemfibrozil: Inhibits both CYP2C9 and the glucuronidation pathway. Can double glimepiride exposure [18].

High-Priority Inducers

  • Rifampin: Reduces sulfonylurea plasma concentrations by 30 to 60%, causing loss of glycemic control within days. Up-titration during rifampin therapy is necessary, with corresponding down-titration after the antibiotic course ends [3].

Drugs That Mask Hypoglycemia

Beta-blockers blunt tachycardia and tremor, leaving diaphoresis as the only reliable adrenergic warning sign. Any patient on a non-selective beta-blocker (propranolol, carvedilol) requires explicit counseling that their hypoglycemia warning symptoms are attenuated [19]. CGM is particularly valuable in this population.


Weight Management During Sulfonylurea Therapy

Sulfonylureas cause an average 2 to 5 kg weight gain over the first 6 to 12 months, driven by increased insulin levels, reduced glycosuria, and hypoglycemia-prompted carbohydrate consumption [20]. Weight gain is not benign in a population already at elevated cardiometabolic risk.

Mitigation Strategies

Options to consider at the 3-month follow-up if weight has increased more than 2 kg [20]:

  1. Add metformin if not already prescribed (neutral to modest weight loss).
  2. Assess whether the patient is treating frequent Level 1 hypoglycemia with excess carbohydrate.
  3. Consider a dose reduction if HbA1c is at or below target, the lowest effective dose carries the lowest hypoglycemia and weight-gain burden.
  4. At 6 months of inadequate glycemic control plus progressive weight gain, discuss transition to a GLP-1 receptor agonist or SGLT2 inhibitor.

Titration Protocols by Agent

Glipizide

Start at 5 mg once daily 30 minutes before the first meal. Up-titrate by 2.5 to 5 mg every 1 to 2 weeks based on fasting glucose. Maximum dose for immediate-release is 40 mg/day in divided doses; for extended-release (Glucotrol XL), 20 mg once daily. Do not exceed 15 mg as a single dose for IR formulation [2][6].

Glimepiride

Start at 1 to 2 mg once daily with the first meal. Titrate by 1 to 2 mg every 1 to 2 weeks. Maximum 8 mg/day. In elderly patients or eGFR 30 to 59, start at 0.5 to 1 mg and do not exceed 4 mg/day without close monitoring [6].

Glyburide

Start at 2.5 to 5 mg once daily with breakfast. Titrate by 2.5 mg weekly. Maximum 20 mg/day. Reserve for patients under 65 with eGFR above 60. This agent should not be the default choice in any new prescription today given the availability of lower-risk alternatives [4][13].


Patient Education Essentials

The following framework (HealthRX SAFE-SU Protocol) summarizes the minimum patient education requirements for anyone starting a sulfonylurea. All five elements should be documented in the visit note.

S, Symptoms of hypoglycemia. Review adrenergic signs (sweating, shaking, palpitations) and neuroglycopenic signs (confusion, slurred speech). Confirm the patient can describe at least two symptoms specific to their experience.

A, Action plan. Instruct 15 g fast carbohydrate, recheck in 15 minutes, repeat if below 70 mg/dL. Provide written instructions. Prescribe glucagon kit for any patient on glyburide or with prior Level 2 event.

F, Food timing. Sulfonylureas require consistent carbohydrate intake at meals. Skipping a meal while taking a morning dose of glyburide or glimepiride creates predictable hypoglycemia within 2 to 4 hours.

E, Exercise awareness. Unplanned physical activity lowers glucose 2 to 4 hours post-exercise. Patients who are physically active need an SMBG check before and after exercise during the first month.

SU, Sick-day rule. Hold the sulfonylurea any day the patient cannot eat normally because of nausea, vomiting, or acute illness. Resume only when oral intake is re-established.


Transitioning Off Sulfonylureas

When to Switch

Transition is appropriate when any of the following develop [4][13]:

  • eGFR falls to the agent-specific threshold
  • Two or more Level 2 hypoglycemia episodes within 3 months
  • HbA1c remains above 8.0% at maximum tolerated dose
  • Weight gain exceeds 5 kg without glycemic benefit

How to Transition

When adding a GLP-1 receptor agonist (e.g., semaglutide, dulaglutide), reduce the sulfonylurea dose by 50% at initiation, then reassess HbA1c at 12 weeks before discontinuing entirely. GLP-1 agents lower glucose through glucose-dependent pathways; overlapping the two classes before the newer agent has reached steady-state plasma concentration carries a real hypoglycemia risk during the transition window [4].

When adding an SGLT2 inhibitor, the same 50% dose-reduction rule applies. The ADA 2024 Standards note that the combination of a sulfonylurea plus an SGLT2 inhibitor roughly doubles the hypoglycemia rate compared to SGLT2 inhibitor plus metformin, based on pooled phase III data [4].


Frequently asked questions

What is the sulfonylureas drug class?
Sulfonylureas are oral insulin secretagogues used in type 2 diabetes. They close ATP-sensitive potassium channels on pancreatic beta cells, forcing insulin release independent of blood glucose level. Second-generation agents (glipizide, glyburide, glimepiride) are standard today and lower HbA1c by 1.0 to 2.0 percentage points.
Which sulfonylurea is safest in chronic kidney disease?
Glipizide is the preferred agent in mild-to-moderate CKD because it is metabolized to inactive compounds by CYP2C9 and does not accumulate in renal impairment. Glyburide must be avoided if eGFR is below 60 mL/min/1.73 m² due to active metabolite accumulation. Glimepiride can be used cautiously at 1 mg/day if eGFR is 30 to 59, but should be avoided below 30.
How often should HbA1c be checked on a sulfonylurea?
Check HbA1c every 3 months during any dose adjustment or when glycemic control is unstable. Once the patient has been at goal for two consecutive readings, extend the interval to every 6 months per ADA 2024 Standards of Care.
What is the biggest risk of sulfonylureas?
Hypoglycemia is the primary safety concern, particularly with glyburide. Unlike newer agents, sulfonylureas release insulin regardless of ambient glucose, so fasting, missed meals, or drug interactions can cause severe hypoglycemia. The ORIGIN trial (N=12,537) documented 5.7 severe hypoglycemia events per 100 patient-years in sulfonylurea-containing regimens.
Can sulfonylureas be used with metformin?
Yes. Combining a sulfonylurea with metformin is a well-established regimen. Metformin reduces baseline insulin resistance while the sulfonylurea augments insulin secretion. The combination achieves greater HbA1c reduction than either agent alone, but the sulfonylurea component still carries full hypoglycemia and weight-gain liability.
Do sulfonylureas cause weight gain?
Yes. Average weight gain is 2 to 5 kg over the first 12 months, driven by increased circulating insulin, reduced glucosuria, and calories consumed to treat hypoglycemia. Patients who experience frequent low glucose episodes tend to gain the most weight. Dose minimization and consistent meal timing are the primary mitigation strategies.
What drugs interact with sulfonylureas?
The most clinically significant interactions are with fluconazole (roughly doubles glipizide AUC via CYP2C9 inhibition), trimethoprim-sulfamethoxazole (hypoglycemia risk within 48 hours), gemfibrozil (doubles glimepiride exposure), and rifampin (reduces sulfonylurea levels by 30 to 60 percent, causing loss of glycemic control). Beta-blockers do not raise sulfonylurea levels but mask adrenergic hypoglycemia symptoms.
What is the maximum dose of glipizide?
For immediate-release glipizide, the maximum is 40 mg/day given in divided doses, with no single dose exceeding 15 mg. For extended-release glipizide (Glucotrol XL), the maximum is 20 mg once daily. Most of the glycemic benefit is achieved at 10 to 20 mg/day; doses above that add hypoglycemia risk without proportionate HbA1c reduction.
Should sulfonylureas be taken with food?
Glipizide immediate-release should be taken 30 minutes before a meal to time peak insulin output with meal-related glucose absorption. Glipizide extended-release, glimepiride, and glyburide are taken with the first meal of the day. Taking a sulfonylurea and then skipping the meal creates a predictable hypoglycemia window 2 to 4 hours later.
Are sulfonylureas safe in elderly patients?
With caution. Older adults have impaired counter-regulatory responses to hypoglycemia, making Level 2 and Level 3 events more dangerous. The American Geriatrics Society Beers Criteria list glyburide as explicitly inappropriate in patients aged 65 and over. Glipizide at low doses is acceptable in elderly patients with preserved renal function and close monitoring.
How long do sulfonylureas take to work?
Fasting glucose begins to fall within the first few days of initiation. Meaningful HbA1c reduction is measurable at 4 to 6 weeks. Full glycemic effect is typically apparent at 3 months, which is why the first formal HbA1c check after starting or titrating is at 3 months.
What is secondary sulfonylurea failure?
Secondary failure is the progressive loss of glycemic efficacy after an initial response, driven by beta-cell exhaustion over time. It occurs in approximately 5 to 10 percent of patients per year on sulfonylurea monotherapy. HbA1c creeping above 8.0% despite maximum tolerated dosing, with no identifiable adherence or dietary explanation, signals secondary failure and the need for regimen intensification.

References

  1. Melander A, Lebovitz HE, Faber OK. Sulfonylureas: why, which and how? Diabetes Care. 1990;13(Suppl 3):18-25. https://pubmed.ncbi.nlm.nih.gov/2209312/
  2. Shorr RI, Ray WA, Daugherty JR, Griffin MR. Individual sulfonylureas and serious hypoglycemia in older people. J Am Geriatr Soc. 1996;44(7):751-755. https://pubmed.ncbi.nlm.nih.gov/8675928/
  3. Niemi M, Backman JT, Neuvonen M, Neuvonen PJ. Effects of gemfibrozil, itraconazole, and their combination on the pharmacokinetics and pharmacodynamics of repaglinide: potentially hazardous interaction between gemfibrozil and repaglinide. Diabetologia. 2003;46(3):347-351. https://pubmed.ncbi.nlm.nih.gov/12687334/
  4. American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes, 2024. Diabetes Care. 2024;47(Suppl 1):S1-S321. https://diabetesjournals.org/care/issue/47/Supplement_1
  5. Inzucchi SE, Lipska KJ, Mayo H, Bailey CJ, McGuire DK. Metformin in patients with type 2 diabetes and kidney disease: a systematic review. JAMA. 2014;312(24):2668-2675. https://pubmed.ncbi.nlm.nih.gov/25536258/
  6. FDA. Glimepiride (Amaryl) Prescribing Information. U.S. Food and Drug Administration. https://www.accessdata.fda.gov/drugsatfda_docs/label/2009/020496s015lbl.pdf
  7. Battelino T, Danne T, Bergenstal RM, 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. https://pubmed.ncbi.nlm.nih.gov/31192680/
  8. Garber AJ, Handelsman Y, Grunberger G, et al. Consensus statement by the American Association of Clinical Endocrinologists and American College of Endocrinology on the comprehensive type 2 diabetes management algorithm, 2020 executive summary. Endocr Pract. 2020;26(1):107-139. https://pubmed.ncbi.nlm.nih.gov/32022600/
  9. Action to Control Cardiovascular Risk in Diabetes Study Group; Gerstein HC, Miller ME, Byington RP, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358(24):2545-2559. https://www.nejm.org/doi/full/10.1056/NEJMoa0802743
  10. Endocrine Society. Clinical Practice Guideline: Hypoglycemia in Adults with Diabetes. J Clin Endocrinol Metab. 2023;108(7):1741-1773. https://pubmed.ncbi.nlm.nih.gov/36802954/
  11. Lipska KJ, Ross JS, Wang Y, et al. National trends in US hospital admissions for hyperglycemia and hypoglycemia among Medicare beneficiaries, 1999-2011. JAMA Intern Med. 2014;174(7):1116-1124. https://pubmed.ncbi.nlm.nih.gov/24838229/
  12. ORIGIN Trial Investigators; Gerstein HC, Bosch J, Dagenais GR, et al. Basal insulin and cardiovascular and other outcomes in dysglycemia. N Engl J Med. 2012;367(4):319-328. https://www.nejm.org/doi/full/10.1056/NEJMoa1203858
  13. Kidney Disease: Improving Global Outcomes (KDIGO) Diabetes Work Group. KDIGO 2022 Clinical Practice Guideline for Diabetes Management in Chronic Kidney Disease. Kidney Int. 2022;102(5S):S1-S127. [https://pubmed.ncbi.nlm.nih.gov/36272764/](https://pubmed.ncbi.nlm.nih.gov/36