ECG criteria
- Measure QT in seconds — from the beginning of the QRS complex to the end of the T wave (preferably in lead II or V5–V6, where the end of T is clearly visible)
- Measure RR in seconds — the interval between adjacent R waves; the choice of lead for RR does not matter
- Apply the formula: QTc = QT + 0.154 × (1 − RR)
- RR and QT are substituted in seconds, and the QTc result is also in seconds (convert to ms by multiplying by 1000 if necessary)
- Compare the obtained QTc with sex-specific normal threshold values
- In tachycardia, Bazett's formula often overestimates QTc, while in bradycardia, it underestimates it; it is advisable to compare the result with Fridericia or Framingham and indicate the formula used. In extreme heart rates, irregular rhythm, wide QRS, or pacing, a single standard formula may be insufficient — clinical interpretation is needed, and sometimes JT/JTc evaluation.
What it looks like on ECG
The Framingham formula is not applied to a single element of the complex, but to the calculation of already measured intervals. On the strip, you need to find:
- Start of QRS — the point of deviation from the isoelectric line before the first wave of the complex.
- End of T wave — the point where the curve returns to the isoelectric line (for low T waves, the tangent method is sometimes used).
- RR — the distance between adjacent R waves.
QT is measured in the lead with the clearest QRS onset and T wave end, often II or V5/V6; RR is measured between successive R waves regardless of the lead. In a regular rhythm, QT is measured and, if necessary, averaged over 3–5 cycles using the same measurement method in each cycle. In atrial fibrillation and other marked rhythm irregularities, several complexes are measured after a standardized number of cycles and a validated averaging method is applied — choosing a single longest QT is not a universal rule and may overestimate the result, since QT also depends on preceding RRs.
Next, QT and RR in seconds are substituted into the formula:
QTc = QT + 0,154 × (1 − RR)
Example: at HR 100 bpm, RR = 0.60 s, measured QT = 0.36 s. QTc = 0.36 + 0.154 × (1 − 0.60) = 0.36 + 0.154 × 0.40 = 0.36 + 0.062 ≈ 0.42 s.
For comparison, Bazett's formula (QTc = QT / √RR) at the same figures gives QTc = 0.36 / √0.60 ≈ 0.46 s — a noticeably higher value, illustrating Bazett's tendency toward overdiagnosis of QT prolongation in tachycardia.
Mechanism
The QT duration reflects the total duration of ventricular depolarization and repolarization. As heart rate increases, diastole shortens more than systole, so QT physiologically shortens during tachycardia and lengthens during bradycardia — regardless of the state of myocardial ion channels. QT correction formulas (Bazett, Fridericia, Framingham, Hodges) attempt to mathematically subtract this HR effect to approximately compare QT duration at different HR values; QTc remains an estimated parameter and is not a direct measurement of repolarization duration. With a wide QRS (bundle branch block, intraventricular conduction delay, ventricular pacing), QT prolongation is partly due to prolonged depolarization itself, rather than repolarization alone — under these conditions, direct application of QTc may falsely indicate repolarization prolongation, and it is advisable to use JT/JTc or a validated QT correction method that accounts for QRS duration. The Framingham formula uses a linear (additive) model of QT dependence on RR, unlike the power models of Bazett (division by the square root of RR) and Fridericia (division by the cube root of RR).
Formula variants for comparison
- Bazett: QTc = QT / √RR — historically the most common, but overestimates QTc in tachycardia and underestimates in bradycardia.
- Fridericia: QTc = QT / RR^(1/3) — power formula, more accurate than Bazett in tachycardia.
- Framingham (Sagie): QTc = QT + 0.154 × (1 − RR) — linear model, RR and QT in seconds.
- Hodges: QTc = QT + 1.75 × (HR − 60), if QT and QTc are in ms (or QTc = QT + 0.00175 × (HR − 60), if QT and QTc are in seconds) — a linear formula that uses HR directly rather than RR.
In a number of population studies, the Framingham formula yielded a lower systematic error in the QT-RR relationship than Bazett's, especially when HR deviated from 60 bpm; there is no universally best formula, and the result should be interpreted taking into account the HR and clinical context. In clinical practice and automated ECG algorithms, Bazett's formula is still used by default due to historical habit, although many guidelines and experts recommend Fridericia or Framingham as alternatives when HR deviates from normal.
When to apply
The Framingham formula is advisable to use:
- in marked tachycardia or bradycardia, when Bazett's formula yields a clearly distorted result;
- during serial monitoring of QTc in the same patient against the background of changing HR (e.g., monitoring therapy with antiarrhythmics or QT-prolonging drugs);
- in scientific and epidemiological studies where the linear model is considered more stable over a wide range of HRs.
In routine ECG interpretation, it is sufficient to know that if formulas disagree at the limit of normal, one should explicitly state which formula was used, and recalculate using the second formula if in doubt.
Common interpretation errors
- Mixing units of measurement: RR and QT must be substituted in seconds, not milliseconds, otherwise the formula result will be grossly incorrect.
- Measuring QT and RR under different conditions (poor recording quality, artifacts), leading to systematic error, especially with varying voltage and T-wave morphology.
- Using the Framingham formula in marked rhythm irregularity (atrial fibrillation) without accounting for the fact that QT depends not only on the current RR but also on preceding intervals — a single measurement does not reflect the true situation.
- Automated QTc calculation by the ECG machine depends on the manufacturer, model, and settings; Bazett's is often historically used, but this must be checked against the machine protocol rather than assumed by default. In case of a borderline or clinically significant result, QT should be measured manually and the correction formula used should be indicated.
- Confusing the end of the T wave with a subsequent U wave — leads to a false overestimation of QT and, consequently, QTc regardless of the correction formula.
1. How is QTc calculated using the Framingham formula?
The Framingham (Sagie) formula is linear (additive): a correction of 0.154 × (1 − RR) is added to the measured QT, where RR is in seconds.
2. How does Bazett's formula typically distort QTc in marked tachycardia compared to linear formulas?
Bazett's power model (division by √RR) typically overestimates QTc in tachycardia more than linear models, although at extreme HRs linear formulas do not guarantee accuracy either.
3. What needs to be done before substituting QT and RR into the Framingham formula?
The Framingham formula is designed for substitution of QT and RR specifically in seconds; using milliseconds without converting the coefficient yields an incorrect result.
Frequently asked questions
How does the Framingham formula differ from Bazett's formula?
Bazett's formula divides QT by the square root of RR (power model), whereas the Framingham formula adds a correction linearly proportional to the deviation of RR from 1 second to QT. In tachycardia and bradycardia, these models yield different results, and Bazett is prone to overestimating QTc at high heart rates.
Which QT correction formula is the most accurate?
There is no single standard: Bazett's formula is simpler and more familiar, but less accurate at extreme HR values; Fridericia's, Framingham's, and Hodges' formulas are often more accurate over a wide HR range, but none is universally best. The choice of formula should be explicitly stated during interpretation, especially at the borderline of normal.
In what units should QT and RR be substituted into the Framingham formula?
In seconds. QT and RR are measured on the ECG in millimeters at a known recording speed and converted into seconds: at 25 mm/s, 1 mm = 40 ms; at 50 mm/s, 1 mm = 20 ms. Only after conversion are the values substituted into the formula: QTc = QT + 0.154 × (1 − RR).
Is it necessary to recalculate QTc manually if the ECG machine has already calculated it?
Yes, if the heart rate differs significantly from normal (marked tachycardia or bradycardia) and the QTc value is borderline normal or pathological. The formula used by the machine depends on its settings and is often unknown without checking the protocol, so in case of a clinically significant QT result, it should be measured manually and the correction formula specified.
Can the Framingham formula be used in atrial fibrillation?
Technically yes, but QT in AF depends not only on the current RR but also on preceding intervals, making correction complex. QT should be measured over several consecutive complexes following a standardized number of cycles and using a validated local method; simply measuring a single cycle or selecting the longest QT may be unreliable.
ECG Course
Interpretation from basics to myocardial infarction — learn to read any ECG using a systematic algorithm.
- Video lessons and theory for every topic
- Simulator case studies under every lesson
- Unlimited ECG simulator practice
- Study guide: "All ECGs in 25 minutes"
ECG Strip Simulator
ECG, radiology, histology, pathomorphology, pharmacology — thousands of flashcards with detailed explanations.
- Real ECG strips, not textbook drawings
- Instant answers and explanations
- Streaks and progress tracking in your dashboard
Related terms
More in the section "Criteria, Formulas, Algorithms"
Sources
- Goldberger's Clinical Electrocardiography, 9th ed.
- Surawicz B., Knilans T. Chou's Electrocardiography in Clinical Practice, 6th ed.
- Orlov V.N. Guide to Electrocardiography
- AHA/ACCF/HRS Recommendations for the Standardization and Interpretation of the Electrocardiogram, 2009
Educational material for medical students and physicians. Does not replace clinical guidelines and physical examination; the ECG must always be interpreted in conjunction with the clinical picture.