Peak Expiratory Flow Predicted Calculator: Four Published Reference Equations, and the Gap Between the European and the Indian Ones

What four published reference equations give for the same adult — Nunn and Gregg’s 1989 UK equation, the NHANES III equation from a United States national survey, and two derived in Indian subjects (a 6,138-adult study using an EU-scale meter, and the Madras series) — with the gap between the European and the Indian prediction printed as the first figure on the page rather than a footnote. For a man of 40 at 170 cm that gap is 156 L/min, or 33%, and it decides what a blow is worth: 400 L/min is 85% of the Indian predicted value and 64% of the European one. The page also prints percentage of personal best, which is what asthma self-management actually uses, and handles the meter scale, which silently changes everything above.

Last updated 6 Oct 2026

This sorts a measurement into a published category. It is not a diagnosis, and one reading taken once is rarely enough to act on. If a result here worries you, take it to a doctor rather than to the internet.

Four published reference equations for the same adult, and how far apart they are

age, height, sex and weight -> a predicted peak expiratory flow from each of four published reference equations, the spread between them, and your own reading as a percentage of each

Sex — every equation on this page is fitted separately for men and womenSex is in every published peak flow equation on this page, and the four of them disagree far more about women than about men. For a man of 40 at 170 cm the four span 472 to 629 L/min, a 33% spread; for a woman of 40 at 155 cm they span 270 to 472 L/min, a 75% spread. Part of that is real and part of it is that the women’s equations were fitted on smaller samples and fit worse: the Madras equation for women has a multiple correlation of 0.275, which is to say it accounts for about 8% of the variation in the women it was fitted on.

UnitsThis applies to the height and weight fields below. Peak flow itself is always in litres per minute on this page, which is what every peak flow meter is marked in. The equations taken from spirometry studies were published in litres per second and are multiplied by 60 here; the formula block below states which ones.

Age in yearsFifteen to eighty. The stated ranges of the four equations are not the same: Nunn and Gregg covers 15 to 85, the Madras series 15 to 63, and NHANES III enrolled from age 8 with a separate equation below 20 for males and below 18 for females, which this page implements rather than extrapolating the adult one down. The age range of the Indian EU-scale study could not be established from the material available here and is recorded as not established rather than guessed. Above 63 the page flags that the Madras equation is being extrapolated. Note the shape Nunn and Gregg’s equation has: because it is fitted on the logarithm of age it rises to a maximum and then falls, and that maximum is at 36.0 years for men and 31.3 for women. The other three fall monotonically from 15.

HeightStanding height without shoes. The page refuses anything outside 130 to 215 cm (51.2 to 84.6 inches) after conversion, which is also what catches the realistic unit mistake: 68 typed into a centimetre field is refused rather than answered, and 170 typed into an inch field becomes 431.8 cm and is refused too. Height matters more than age in every one of these equations. On the Indian EU-scale equation a centimetre of height is worth 3.21 L/min in a man, against 1.81 L/min for a year of age — so a measurement error of two centimetres moves the prediction as much as three and a half years of ageing.

Weight — used by one of the four equations onlyOnly the Madras equation uses weight; the other three do not, and their predictions do not change when you change this field. It is on the page because the Madras equation as published includes a weight term, and dropping the term to make the input set tidier would be using a different equation from the one that was published. In a man the weight coefficient is 0.054 L/s per kilogram, so ten kilograms is worth 32 L/min — which is a large effect for a variable the other three equations leave out altogether.

Which scale your own meter is marked inThis changes nothing in the arithmetic and a great deal in what the arithmetic means. The original Wright scale was found to overestimate peak flow in the mid-range by up to 80 L/min and to underestimate above about 650, so the UK and EU moved to a linear scale in 2004. The same blow reading 440 on a Wright-scale meter reads about 387 on an EU-scale one. Nunn and Gregg’s equation was fitted on Wright-scale readings; the Indian EU-scale equation was fitted on EU-scale readings; the NHANES III and Madras equations came from spirometers and match neither meter. So part of the gap between the European and the Indian prediction on this page is the scale and not the population, and this page says so rather than pretending otherwise. It does not convert between the scales: the published conversion is a non-linear table and reproducing or approximating it is out of scope here.

Your own best reading, if you have measured one — 0 for noneThe best of three blows, which is how a peak flow is taken. Leave it at 0 and every row that uses it is hidden. The page accepts 50 to 900 L/min and ignores anything outside that. Peak flow is the most effort-dependent measurement in routine respiratory practice: a half-hearted blow, a tongue in the mouthpiece, a leak at the lips or a slow start all give a low reading on healthy lungs, and nothing on this page or on any calculator can tell a low reading from bad technique apart from a low reading from narrowed airways.

Your personal best, if you have one — 0 for noneA personal best is established by taking a peak flow every day for two to three weeks while asthma is under control, in the National Institutes of Health patient materials between noon and 2 pm and after a reliever, and taking the highest. Where a personal best exists it replaces every predicted value on this page. The percentages written on an asthma action plan are percentages of a personal best, not of a population equation, and that is exactly why: a personal best needs no equation, no ethnic reference and no view about which scale the equation was fitted on.

Which equation gives the headline figureThis chooses which equation the headline number comes from. All four are computed and printed in the rows below whichever you choose, because the disagreement between them is the most useful thing on this page. They are not variants of one method: different populations, different decades, different instruments and different scales. The default is the Indian EU-scale equation because it is the only one of the four fitted on Indian subjects using the scale that every meter sold today carries, which makes it the one that matches most readers of this page on both counts at once. Choosing it is not the same as saying it is right.

473L/minExample

a man of 40 years, 170 cm and 65 kg, headline from the Indian EU-scale equation

The four equations, their populations, their instruments and their scales

Indian adults, EU-scale meter (Kodgule 2014): PEF = 3.206H − 1.807A for men; PEF = 2.368H − 1.454A for women. Litres per minute, no intercept. · UK adults, Wright-scale meter (Nunn and Gregg 1989): ln(PEF) = 0.544 ln(A) − 0.0151A − 74.7/H + 5.48 for men; ln(PEF) = 0.376 ln(A) − 0.012A − 58.8/H + 5.63 for women. Litres per minute. · White US adults, spirometer (NHANES III, Hankinson 1999): PEF = b0 + b1A + b2A² + b3H², litres per second, ×60 here. Men 20 and over: 1.0523 + 0.08272A − 0.001301A² + 0.00024962H²; men under 20: −0.5962 − 0.12357A + 0.013135A² + 0.00024962H². Women 18 and over: 0.9267 + 0.06929A − 0.001031A² + 0.00018623H²; women under 18: −3.6181 + 0.60644A − 0.016846A² + 0.00018623H². Its published fifth-percentile lower limit of normal uses the same equation with a smaller H² coefficient — 0.00017635 for men, 0.00012148 for women — so LLN = PEF − 0.00007327H² in men and PEF − 0.00006475H² in women, before the ×60. · South Indian adults, spirometer (Madras, Vijayan 1993): PEF = −5.837 − 0.017A + 0.064H + 0.054W for men; PEF = −0.179 − 0.006A + 0.025H + 0.021W for women. Litres per second, ×60 here. · spread = max − min, spread% = 100(max − min)/min, gap = Nunn and Gregg − Kodgule, % predicted = 100 × your reading / predicted, % personal best = 100 × your reading / your best.

A: age in years. The only equation on this page that is not monotonically falling in age above 15 is Nunn and Gregg’s: because it carries both ln(A) and A, it has a maximum at A = 0.544/0.0151 = 36.0 years in men and 0.376/0.012 = 31.3 years in women
H: standing height in centimetres. Height is the dominant term in all four. In the Indian EU-scale equation a centimetre is worth 3.206 L/min in a man, and a year of age 1.807, so two centimetres of measurement error equals three and a half years of ageing
W: weight in kilograms, used by the Madras equation alone. The other three do not take weight and their predictions do not move when it changes
PEF: a PREDICTED peak expiratory flow for a population average of that age, height and sex. It is not a measurement, it is not a threshold, and no asthma severity or diagnosis is derived from it anywhere on this page
LLN: the lower limit of normal, meaning the fifth percentile of the reference population: 5% of healthy subjects in NHANES III fell below it. It is published only for the NHANES III equation. For a man of 40 at 170 cm it is 127 L/min, or 22.3%, below that equation’s own predicted value
the scales: Nunn and Gregg’s readings came from the original Wright-scale meter, which overestimates in the mid-range by up to 80 L/min and underestimates above about 650; the EU scale (EN 13826, now ISO 23747) replaced it in 2004 and the Indian equation was fitted on it. NHANES III and the Madras series used spirometers. This page matches equations to scales in its notes and does NOT convert between the scales
why no conversion: the Wright-to-EU conversion published after the 2004 changeover is a non-linear table, distributed by the meter manufacturer. Reproducing it, redrawing it or fitting a curve through it would be reproducing a copyrighted table and would also put a number on the page that looks like a measurement. The page names the difference, gives its published size, and leaves the conversion to the manufacturer’s own chart

Worked example

a man of 40 years, 170 cm and 65 kg, headline from the Indian EU-scale equation
All four, for the same man. Age 40 years, height 170 cm, weight 65 kg. Indian EU-scale equation: 3.206 × 170 − 1.807 × 40 = 545.02 − 72.28 = 472.7 L/min. Nunn and Gregg: exp(0.544 ln 40 − 0.0151 × 40 − 74.7/170 + 5.48) = exp(2.0068 − 0.6040 − 0.4394 + 5.48) = exp(6.4434) = 628.5 L/min. NHANES III, white men 20 and over: (1.0523 + 0.08272 × 40 − 0.001301 × 1600 + 0.00024962 × 28900) × 60 = 9.4935 × 60 = 569.6 L/min. Madras: (−5.837 − 0.017 × 40 + 0.064 × 170 + 0.054 × 65) × 60 = 7.873 × 60 = 472.4 L/min. The headline is the Indian EU-scale equation, so 473 L/min.
The gap, which is the point of the page. The European equation gives 628.5 and the Indian one 472.7, so the gap is 155.8 L/min, which is 32.9% of the Indian figure. The 2014 study that fitted the Indian equation reported exactly this in general terms: an Indian adult of average height and age has a peak flow about 30% below the corresponding European adult on Nunn and Gregg's equation. The full spread across all four is 472.4 to 628.5 L/min, 156 L/min or 33.1% of the lowest. For a woman of 40 at 155 cm and 50 kg the same four span 270 to 472 L/min, a 74.7% spread, and the European-to-Indian gap alone is 52.9%.
What that does to a blow. Take a reading of 400 L/min from this man on an EU-scale meter. Against the Indian equation it is 84.6% of predicted; against Nunn and Gregg it is 63.6%; against NHANES III 70.2% and against Madras 84.7%. The conventional line at 80% of predicted falls at 378 L/min on the Indian equation and at 503 L/min on the European one — a 125 L/min difference in where the same convention sits. A blow of 500 L/min is 105.8% of the Indian predicted value and 79.6% of the European one, so the same man is simultaneously above his predicted value and below 80% of predicted depending on which published equation you divide by. That is not a subtlety; it is the difference between two zones on a plan built the wrong way.
The two Indian equations, and why their agreement matters. At exactly 170 cm they differ by 0.4 L/min — 472.7 against 472.4 — which is a crossing point and not general agreement: at 150 cm the EU-scale equation sits 13.0 L/min above the Madras one and at 190 cm 12.3 L/min below it. But the scale of the disagreement is the thing. Two Indian studies twenty-one years apart, one with a laboratory spirometer on 273 adults in Madras and one with an EU-scale meter on 6,138 adults across the country, land within about 13 L/min of each other across the whole adult height range at age 40, while both sit roughly a third below the European equation. They do separate with age, because the EU-scale equation falls by 1.81 L/min a year and the Madras one by only 1.02: at 170 cm the EU-scale figure is 20 L/min above the Madras one at age 15 and 31 L/min below it at 80. That is still small beside the European gap, and instrument and era do not explain a third.
The interval nobody prints. NHANES III is the only one of the four that publishes a lower limit of normal, and it is the number that should sit beside every percentage of predicted. For this man the NHANES III prediction is 569.6 L/min and its published fifth percentile is 442.6 — 127.1 L/min, or 22.3%, below its own predicted value. In other words, in the population that equation was fitted on, one healthy adult in twenty blows more than 22% under predicted. The chart adapted from Nunn and Gregg says the same thing in different units: in men, readings up to 100 L/min below predicted are within normal limits, and in women up to 85. Any reading of percentage-of-predicted that treats 85% as abnormal is ignoring both of those published statements.
Why personal best beats all of this. The asthma action plan template published by the United States National Institutes of Health is written entirely in percentages of a personal best: more than 80% of your best in the first zone, 50% to 79% in the second, under 50% in the third (NIH Publication No. 20-HL-5251, February 2021). A personal best is the highest of daily readings taken over two to three weeks while asthma is under control. It needs no reference population, no ethnic correction and no view about which meter scale the equation was fitted on, and it is measured on the reader's own meter with the reader's own technique. Where a personal best exists, it replaces every predicted figure on this page. What this page will not do is tell anybody what to do at any of those percentages: the zone numbers that matter are the ones written on a reader's own plan, by the person who wrote it.
And what the measurement cannot do. Peak flow is the most effort-dependent routine respiratory measurement there is — a slow start, a leak at the lips, a tongue over the mouthpiece or a half-hearted blow all produce a low number from healthy lungs, and no calculation can distinguish that from narrowed airways. It is also a poor proxy for the spirometric measurement it is often used to stand in for: in a study at an Indian teaching hospital, peak flow percent predicted and forced expiratory volume percent predicted were more than five percentage points apart in about three quarters of patients with airflow obstruction. A low figure here is a reason to see a doctor, not a diagnosis, and the right instrument for a diagnosis is a spirometer.

The four equations across adult life, for a man of 170 cm and 65 kg

Read the last two columns. The four published equations are never closer than 16% of the lowest of them and at age 50 they are 34% apart, and almost all of that is the step down from the two European-derived equations to the two Indian-derived ones. The European minus Indian gap widens from 20 L/min at 15 years to about 158 L/min in the mid-forties and then narrows again, because Nunn and Gregg’s equation is fitted on the logarithm of age and so rises to a maximum at 36 before falling, while the Indian equation falls in a straight line throughout. The Madras figures above 63 years are extrapolations beyond that equation’s stated range of 15 to 63 and are shown so the extrapolation can be seen rather than hidden: its very shallow age slope makes it the highest of the Indian figures in older men, which is an artefact of the extrapolation and not a finding. Figures are rounded to whole litres per minute, which is finer than any hand-held meter can be read.
Age	Indian, EU scale (Kodgule)	UK, Wright scale (Nunn and Gregg)	White US, spirometer (NHANES III)	South Indian, spirometer (Madras)	Lowest to highest	Spread	European minus Indian
15	518	538	463	498	463 to 538	75 L/min (16.1%)	20 L/min (3.8%)
20	509	583	564	493	493 to 583	90 L/min (18.3%)	74 L/min (14.6%)
30	491	625	575	483	483 to 625	142 L/min (29.5%)	134 L/min (27.4%)
40	473	629	570	472	472 to 629	156 L/min (33.1%)	156 L/min (32.9%)
50	455	610	549	462	455 to 610	155 L/min (34.2%)	155 L/min (34.2%)
60	437	579	513	452	437 to 579	143 L/min (32.7%)	143 L/min (32.7%)
70	419	542	461	442	419 to 542	123 L/min (29.4%)	123 L/min (29.4%)
80	400	501	393	432	393 to 501	107 L/min (27.3%)	100 L/min (25.1%)

The same four equations for a woman of 155 cm and 50 kg

The disagreement in women is roughly twice what it is in men: the four equations span 51% to 75% of the lowest at every age, and the European minus Indian gap alone reaches 55%. Two things drive it. The gap between the European and Indian equations really is larger in women. And the women’s equations are the weaker fits — the Madras equation for women has a multiple correlation of 0.275 against 0.553 for men, and NHANES III reports a coefficient of determination of 0.5559 for white women against 0.7808 for white men — so the Madras column in particular is close to quoting a group mean. It is shown with its fit statistic rather than left out, because a reader who finds it on another site should be able to see what it is worth.
Age	Indian, EU scale (Kodgule)	UK, Wright scale (Nunn and Gregg)	White US, spirometer (NHANES III)	South Indian, spirometer (Madras)	Lowest to highest	Spread	European minus Indian
15	345	441	370	279	279 to 441	162 L/min (57.8%)	96 L/min (27.7%)
20	338	463	382	278	278 to 463	185 L/min (66.7%)	125 L/min (36.9%)
30	323	478	393	274	274 to 478	204 L/min (74.5%)	155 L/min (47.8%)
40	309	472	391	270	270 to 472	202 L/min (74.7%)	163 L/min (52.9%)
50	294	456	377	267	267 to 456	189 L/min (70.8%)	161 L/min (54.8%)
60	280	433	351	263	263 to 433	170 L/min (64.4%)	153 L/min (54.7%)
70	265	407	312	260	260 to 407	147 L/min (56.7%)	141 L/min (53.3%)
80	251	379	261	256	251 to 379	129 L/min (51.3%)	129 L/min (51.3%)

One blow, five percentages: a man of 40 at 170 cm and 65 kg on an EU-scale meter

This is the table that makes the point concrete. The same reading is worth about 21 percentage points more against the Indian equation than against the European one for this man, and that decides which side of a conventional line it falls on. Eighty per cent of the Indian predicted value is 378 L/min; eighty per cent of Nunn and Gregg’s is 503. A blow of 500 is 105.8% of the Indian figure and 79.6% of the European one — above predicted on one equation and below the conventional 80% line on the other, for the same man on the same meter in the same minute. Note also that the Indian and Madras columns are nearly identical here, because the two equations happen to cross at about 170 cm in men; they are not this close at other heights.
Reading (L/min)	As % of the Indian EU-scale equation	As % of the Madras equation	As % of NHANES III	As % of Nunn and Gregg
250	52.9%	52.9%	43.9%	39.8%
300	63.5%	63.5%	52.7%	47.7%
350	74.0%	74.1%	61.4%	55.7%
400	84.6%	84.7%	70.2%	63.6%
450	95.2%	95.3%	79.0%	71.6%
500	105.8%	105.8%	87.8%	79.6%

What a peak flow meter scale change does, and why this page does not convert between scales

The meter scale is the quiet error on pages like this one. A reader in 2026 almost certainly has an EU-scale meter and almost certainly finds a chart derived from Wright-scale data, and the difference runs in the same direction as the population difference — both make the reading look worse. Stacked, on a man of 40 at 170 cm, they are the distance between 85% of predicted and 64%.
Question	What the published material says	What this page does
Which scale is my meter?	Meters sold in the UK and EU since 2004 are marked to EN 13826, now carried forward as ISO 23747, and are referred to as EU scale. Anything older is on the original Wright scale.	Asks, and uses the answer only to say which equations are scale-matched to the reader. The arithmetic does not change.
How big is the difference?	The original Wright scale was found to overestimate peak flow in the mid-range by up to 80 L/min and to underestimate above about 650. A blow reading 440 on a Wright-scale meter reads about 387 on an EU-scale one. The underlying measurement work is Miller, Dickinson and Hitchings, Thorax 1992;47:904–9.	States the size and direction. Does not apply it to any number.
Can I convert my reading?	A conversion between the two scales exists and is distributed as a chart and an online converter by the meter manufacturer. It is non-linear, so there is no single factor.	No conversion is offered. The conversion is a copyrighted table; reproducing it, redrawing it or fitting a curve through it would reproduce the table in substance, and would also put a converted figure on the page that would read like a measurement.
Which equation matches my meter?	Nunn and Gregg 1989 was fitted on Wright-scale readings. The Indian equation of 2014 was fitted on an EU-scale meter. NHANES III 1999 and the Madras series 1993 used laboratory spirometers.	Says which, in a note that changes with the scale you select. Scale-matching is part of why the page prints four numbers instead of one.
Does the predicted-value chart on my diary account for it?	The normal-values chart in common use states that it is adapted by the manufacturer for use with EU-scale meters from Nunn and Gregg’s 1989 equation, and that its values are derived from Caucasian populations. It gives the equation’s normal limits as up to 100 L/min below predicted in men and 85 in women, for ages 15 to 85, men 160 to 190 cm and women 152 to 183 cm.	Uses Nunn and Gregg’s equation as published, not the manufacturer’s adapted chart, and says so. The chart itself is not reproduced. Its stated normal limits are quoted because a limit is a fact; the nomogram is not redrawn.
Does any of this change what I should do?	Nothing in the scale literature is advice about treatment.	Nothing on this page is either. It prints numbers and names their sources.

Where each figure on this page comes from, and what was deliberately left out

Two decisions shape this page. A published regression is a method and its coefficients are facts about data, so the four equations are computed freely even where the paper carrying one is not openly readable; a manufacturer’s nomogram and a manufacturer’s scale-conversion table are expression, and neither is reproduced, redrawn or approximated. And where something could not be established — the Indian study’s full author list, age range and residual error, a first-hand reading of Nunn and Gregg, the units printed in the NHANES III tables, the provenance of the fifth formula — the page says so in the place where it would have gone, rather than substituting something that looks like an answer.
Figure	Source used	What was rejected, or could not be established
Indian EU-scale equation	Kodgule R, et al. Reference values for peak expiratory flow in Indian adult population using a European Union scale peak flow meter. J Postgrad Med 2014;60(2):123–9, doi:10.4103/0022-3859.132311. Multicentre, 6,138 healthy Indian adults (3,720 men, 2,418 women) measured on an EU-scale meter; equations derived with gender, age and height as the significant determinants and validated on a fifth of the sample with mean predicted-minus-measured differences of 1.85 L/min in men and 1.64 L/min in women. Source also of the statement that an Indian adult of average height and age has a peak flow about 30% lower than the corresponding European adult on Nunn and Gregg’s equation.	The full author list, the age range and the residual standard deviation. The journal’s own article page could not be read from here, so the citation is given as the first author and et al. rather than inventing an order, and the age range and residual error are recorded as not established. The study first appeared as an oral presentation at the 2011 European Respiratory Society congress under the title Derivation of a predicted equation for peak expiratory flow (PEF) values in adult Indian population using EU scale peak flow meter (PFM), with twelve named authors, and the instrument is described there and in the open abstract as a Breathometer (Cipla Ltd, India) marked to the EU scale.
Nunn and Gregg equation	Nunn AJ, Gregg I. New regression equations for predicting peak expiratory flow in adults. BMJ 1989;298(6680):1068–70, PMID 2497892. Predicted values derived for men and women aged 15 to 85.	First-hand reading of the paper. The full text on the open archive was behind an automated challenge and could not be read from here. The coefficients used (men 0.544, 0.0151, 74.7, 5.48; women 0.376, 0.012, 58.8, 5.63, with height in centimetres and the result in litres per minute) are taken from an independent clinical calculator that cites the paper directly, and were checked three ways before being used: the implied age of maximum predicted flow is 36.0 years in men and 31.3 in women, which matches the rise-then-fall shape the paper’s model was built to produce; the values fall in the right place against the published normal-values chart’s stated age range of 15 to 85; and height in metres rather than centimetres makes the exponential collapse to near zero, which fixes the unit that one secondary source states wrongly as metres. Any residual risk is in the fourth significant figure of the coefficients, not in the structure.
NHANES III equation and its lower limit of normal	Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med 1999;159:179–87, read from the copy hosted by the United States Centers for Disease Control and Prevention (CDC Stacks 198173). PEF equations of the form b0 + b1·age + b2·age² + b3·height², with separate coefficients for the predicted value and for the fifth-percentile lower limit of normal differing only in the height² term.	Any ethnic group an Indian reader belongs to. NHANES III reports equations for white, African-American and Mexican-American subjects and includes no South Asian or Indian group, which is stated on the page rather than papered over by borrowing the white equation. Also not asserted: the units of PEF in the paper’s own tables, which are not printed in them. Litres per second is used here and was confirmed arithmetically — the alternative gives physically impossible values — and the two age branches were confirmed by checking that they meet at the switch, 9.4004 against 9.4003 L/s at age 20 for men and 6.6072 against 6.6074 at age 18 for women.
Madras equation	Vijayan VK, Kuppurao KV, Venkatesan P, Sankaran K. Reference values and prediction equations for maximal expiratory flow rates in non-smoking normal subjects in Madras. Indian J Physiol Pharmacol 1993;37(4):291–7. 273 healthy non-smoking adults aged 15 to 63 (144 men, mean age 29.9 years, height 166.1 cm, weight 56.0 kg; 129 women, mean age 30.1, height 151.8 cm, weight 45.7 kg), measured on a Transfer Test Model C spirometer. PEFR equations with age, height and weight terms; multiple correlation 0.553 and standard error 1.574 L/s in men, 0.275 and 1.013 L/s in women.	Nothing rejected, but two things stated rather than smoothed over. It is a spirometric measurement, not a hand-held meter reading. And the women’s equation accounts for about 8% of the variation in its own derivation sample, which the page prints beside it. The paper’s own comparison with Caucasian values is qualitative — flow rates at lower lung volumes in men similar to Caucasians, in women lower than Caucasians — and no percentage from it is quoted here, because none is given.
The 80%, 50% and personal-best convention	The asthma action plan template published by the United States National Heart, Lung, and Blood Institute, NIH Publication No. 20-HL-5251, February 2021, which is written as peak flow more than 80% of my best in the green zone, 50% to 79% of my best in the yellow, and less than 50% of my best in the red. The method for establishing a personal best — a daily peak flow for two to three weeks with asthma under control, between noon and 2 pm and after a reliever, taking the highest — is from the United States National Library of Medicine’s MedlinePlus patient instructions.	Any zone assignment, and any action. The page prints 80% and 50% of a personal best as arithmetic and names where the convention comes from. It does not label a reading with a zone and it does not say what to do at any percentage, because a real action plan is written in litres per minute by the clinician who wrote it, may not use the default percentages, and attaches actions specific to one person’s medicines. United States Government works are used here in preference to the equivalent material from other bodies, which is this project’s standing position where a figure exists in more than one place.
Peak flow meter scales	The size and direction of the Wright-to-EU difference — overestimation in the mid-range by up to 80 L/min, underestimation above about 650, and 440 on the Wright scale reading about 387 on the EU scale — traces to Miller MR, Dickinson SA, Hitchings DJ. The accuracy of portable peak flow meters. Thorax 1992;47:904–9, and is quoted here from the occupational asthma reference material that cites it. The standard itself is EN 13826, carried forward as ISO 23747.	The conversion between the scales, in every form. It is published as a non-linear chart and an online converter by the meter manufacturer; it is not reproduced, redrawn, tabulated or fitted here. Also not reproduced: the manufacturer’s EU-scale normal-values nomogram adapted from Nunn and Gregg. Its stated normal limits are quoted — up to 100 L/min below predicted in men, 85 in women, for ages 15 to 85, men 160 to 190 cm and women 152 to 183 cm, with its own note that the values are derived from Caucasian populations — because a stated limit is a fact, while the nomogram is expression.
The disagreement between published formulae in general	Radeos MS, Camargo CA Jr. Predicted peak expiratory flow: differences across formulae in the literature. Am J Emerg Med 2004;22(7):516–21. Compared the published formulae (Godfrey 1970, Polgar and Promhadat 1971, Gregg and Nunn 1973, Leiner 1963, Hankinson 1999, Hsu 1979) and found that an 18-year-old man’s 100%-predicted value ranged from 501 to 730 L/min, a spread of 229 L/min, and that a 35-year-old woman could be graded severe at 46% of predicted on one formula and moderate at 57% on another.	Nothing. This is the paper that makes the case for the way this page is built, and it is cited for that rather than for any coefficient.
A fifth equation widely implemented in software	Not used. A linear formula appears in United Kingdom general-practice software and in several online calculators — for men, PEF = (5.48 × height in metres + 1.58 − 0.041 × age) × 60, and for women (3.72 × height + 2.24 − 0.03 × age) × 60 — giving 582 L/min for a man of 40 at 1.78 m.	Its provenance. One widely circulated chart attributes it to a 2004 review that is itself a comparison of other people’s formulae; another reference source prints it with the source explicitly unspecified; and it is plausibly but not verifiably Leiner’s 1963 equation, which that same review does cite. Rather than attach a citation that could not be confirmed, the formula is named here and not computed. A reader who finds a different predicted value on another site is quite likely looking at this.
A single modern multi-ethnic equation	Does not exist for peak flow. The Global Lung Function Initiative’s 2012 reference equations brought multi-ethnic reference values to spirometry, and the Initiative states plainly that data on PEF and other instantaneous flows were not collected for GLI-2012, so manufacturers cannot depict predicted contours for them.	Nothing to reject. It is recorded because it explains why this page has four single-population equations instead of one equation with an ethnic coefficient, and why that is not a shortcoming of the page.
Anything expressed as a diagnosis, a severity grade or a treatment	Nothing. The page has no diagnostic criterion, no severity grade and no treatment anywhere in it, by design.	Every form of it. Asthma severity grading from peak flow percent predicted exists in the guideline literature and is deliberately absent here: it would require this page to choose one reference equation, which is the single thing it is built to avoid. One supporting finding is quoted instead — from a study at an Indian teaching hospital, peak flow percent predicted and forced expiratory volume percent predicted were more than five percentage points apart in roughly three quarters of patients with airflow obstruction — to show that the percentage this page prints is not interchangeable with the measurement a severity grade is usually based on.

Why this page prints four predicted values and the gap between them instead of one number

A predicted peak flow is a statement about a population, and the population matters more than most pages admit. Reference equations fitted in European populations predict substantially higher peak flows than equations fitted in Indian subjects at the same age, height and sex. The 2014 study that fitted the equation this page defaults to measured 6,138 healthy Indian adults on an EU-scale meter and reported that an Indian adult of average height and age has a peak flow about 30% lower than the corresponding European adult on Nunn and Gregg’s equation. This page computes the difference rather than describing it: for a man of 40 at 170 cm it is 156 L/min, which is 32.9% of the Indian figure and 24.8% of the European one; for a woman of 40 at 155 cm it is 163 L/min, 52.9% of the Indian figure and 34.6% of the European one. Both denominators are printed above, because they are not interchangeable and the difference between them is large: the study’s own phrasing, about 30% lower than the European adult, is a percentage of the European value, and this page reproduces it at 25.1% for a man of 35 at 165 cm and 34.7% for a woman of 35 at 152 cm, which brackets the stated figure. A reader who takes a blow on a modern meter and compares it against the chart on the back of a peak flow diary is comparing it against a Caucasian equation, and the chart in common use says so in its own footnote. The consequence is not academic: the same blow of 400 L/min from that man is 84.6% of the Indian predicted value and 63.6% of the European one.

The four equations are not variants of one method and they should not be averaged. They differ in population, decade, instrument and scale, and in the shape of the model. Nunn and Gregg’s 1989 equation is fitted on the logarithm of age as well as age, so the predicted value rises to a maximum and then falls — at 170 cm it takes a man from 538 L/min at 15 up to 630 at 35 and back down to 501 at 80, with the turning point at 36.0 years. The Indian EU-scale equation is a straight line in age with no intercept at all. NHANES III is quadratic in age and linear in height squared. The Madras equation is linear in age, height and weight, and is the only one of the four that uses weight. Because the shapes differ, the gap between them is not a constant that could be corrected away with a factor: in men it widens from about 4% at age 15 to 34% at 50 and narrows to 25% at 80.

The meter scale is the second error, it runs in the same direction as the first, and almost nothing mentions it. The original Wright scale was found to overestimate peak flow in the mid-range by up to 80 L/min and to underestimate above about 650, which is why the UK and EU moved in 2004 to the linear scale defined by EN 13826 and now carried as ISO 23747. A blow reading 440 on a Wright-scale meter reads about 387 on an EU-scale one. Nunn and Gregg’s equation was fitted on Wright-scale readings; the Indian equation was fitted on EU-scale readings; NHANES III and the Madras series came from laboratory spirometers and match neither meter. So a reader with a meter bought in the last twenty years, comparing against a chart built from Nunn and Gregg, is penalised twice over, and part of the 33% gap this page prints is scale rather than population. Nobody has published a clean separation of the two, and this page says that rather than assigning the gap to one cause. It also does not convert between the scales: the published conversion is a non-linear table distributed by the meter manufacturer, and reproducing or approximating it is out of scope here for the same reason the paediatric weight page does not redraw a length tape.

Percentage of predicted is the weaker of the two percentages, and this page says which is which. The asthma action plan template published by the United States National Institutes of Health is written entirely in percentages of a personal best — more than 80% of your best, 50% to 79%, under 50% — and not in percentages of a predicted value. That choice is the answer to everything above. A personal best is the highest of daily readings taken over two to three weeks while asthma is under control, measured on the reader’s own meter with the reader’s own technique, so it carries no reference population, no ethnic correction and no assumption about the scale. Where a personal best exists it replaces every predicted figure on this page. Predicted values have a use — they are what you have when somebody presents for the first time and has never had a personal best — but they are the fallback and not the standard. What this page will not do is issue a plan. It prints 80% and 50% of a personal best as arithmetic and explains where that convention comes from, and it says nothing about what to do at any of those numbers, because the numbers that matter are the ones written on a reader’s own plan and the actions beside them were chosen by the person who wrote it.

The interval is the part everybody drops. Of the four equations here, only NHANES III publishes a lower limit of normal, and it is startling: for a man of 40 at 170 cm its predicted value is 570 L/min and its own published fifth percentile is 443, which is 22.3% below predicted. One healthy adult in twenty in that reference population blows more than 22% under the predicted value. The chart adapted from Nunn and Gregg states the same fact in absolute units: in men, readings up to 100 L/min below predicted are within normal limits, and in women up to 85. Set those beside the habit of treating 80% of predicted as a threshold and the habit does not survive. The Madras equation makes the point from the other end, through its own fit statistics: a multiple correlation of 0.553 in men with a standard error of 1.574 L/s, which is 94 L/min, and 0.275 in women with a standard error of 1.013 L/s, which is 61. An equation accounting for about 8% of the variation among the women it was fitted on is close to quoting their mean, and that is stated on this page rather than hidden behind a tidy number.

What the measurement itself cannot do, which no amount of better arithmetic fixes. Peak flow is the most effort-dependent measurement in routine respiratory practice. A slow start, a leak at the lips, a tongue over the mouthpiece or simply a half-hearted blow all produce a low figure from entirely healthy lungs, and neither this page nor any other can distinguish that from narrowed airways. Peak flow also agrees poorly with the spirometric measurement it is often used to stand in for: in a study at an Indian teaching hospital, peak flow as a percentage of predicted and forced expiratory volume in one second as a percentage of predicted were more than five percentage points apart in roughly three quarters of patients with airflow obstruction. And a low peak flow has many causes besides asthma. So the page predicts and compares, and stops: no diagnosis, no severity grade, no treatment, and no suggestion that any figure here settles anything. If a reading worries you, the next step is a doctor with a spirometer.

What else is on this site, and what deliberately is not. Exertion and cardiorespiratory fitness are the VO2 max page and the training zones page; cumulative tobacco exposure, which is the other thing a reader with a low blow usually wants to quantify, is the pack-years page. There is no spirometry interpretation page here and that is deliberate: forced expiratory volume, forced vital capacity and their ratio are a different measurement with their own reference equations and their own obstruction criteria, and a page that mixed a hand-held meter reading into them would invite exactly the substitution the literature warns against. Paediatric peak flow is also not on this page: the childhood equations are height-only, there is a widely used arithmetic approximation that this page will not reproduce because its provenance could not be established, and children are a separate reference problem. Weight estimation in children, for comparison, is the paediatric weight page, which is built on the same principle as this one: print every published formula and the distance between them, rather than one number that hides it.

Frequently asked questions

Which of the four equations should I use?

If you have a personal best, none of them — use the personal best, and read the percentage of it instead. If you do not, then for an Indian adult with a meter bought in the last twenty years the Indian EU-scale equation is the only one of the four that matches both the population and the scale, which is why it is the default. That is not the same as saying it is right: its age range could not be established from the material available here, it has no intercept so it extrapolates towards zero in a straight line, and it has not been independently validated outside its own study as far as could be established. The honest reading of this page is the spread, not any single row. And if a clinician has already told you which predicted value they use, use theirs, so that your figure and theirs are the same number.

Why is my reading so much lower than the chart on my peak flow diary says it should be?

Two reasons stack, and both run in the same direction. The chart in common use is adapted from Nunn and Gregg’s 1989 equation and states in its own footnote that its values are derived from Caucasian populations; in Indian adults the published difference is about 30% at average height and age. And the equation behind it was fitted on readings from the old Wright-scale meter, which overestimates in the mid-range by up to 80 L/min relative to the EU-scale meter you almost certainly have. Together, for a man of 40 at 170 cm, that is the difference between 85% of predicted and 64%. There is a third reason worth checking before any of that, which is technique: a slow or half-hearted blow gives a low number from healthy lungs. None of this means a low reading does not matter — it means a percentage of predicted from a Western chart is not the evidence it looks like, and a personal best taken on your own meter is.

What are the green, amber and red zones, and will this page tell me which one I am in?

No, and the reason is not squeamishness. The zones are a convention in asthma self-management and the usual percentages are published: the asthma action plan template from the United States National Institutes of Health is written as more than 80% of your best peak flow, 50% to 79% of your best, and under 50% of your best (NIH Publication No. 20-HL-5251, February 2021). But the numbers on a person’s plan are written on that plan, by the clinician who wrote it, in litres per minute and not in percentages, and they are often not the default percentages — a plan may use different cut points for a particular person, and the actions beside each zone are specific to that person’s medicines. A calculator that printed a zone would be issuing a plan it knows nothing about. So this page prints 80% and 50% of whatever personal best you enter, as arithmetic, explains where the convention comes from, and stops there.

How do I work out a personal best?

The patient materials published by the United States National Library of Medicine describe it as taking a peak flow every day for two to three weeks while asthma is under control, between noon and 2 pm and after a reliever, and taking the highest figure. The important details are that it is your meter, your technique and your controlled state: a personal best from somebody else’s meter is not comparable, and a personal best established during a bad patch is too low. It also goes stale. A growing young adult’s best rises; after a long period of poor control, and after treatment that improves control, the figure needs re-establishing. Peak flow is taken as the best of three blows, standing, with a sharp short maximal effort rather than a long one.

Why does the page ask for my weight when three of the four equations ignore it?

Because the Madras equation as published includes a weight term, and dropping the term to make the input set tidier would mean computing a different equation from the one the paper reports. Its weight coefficient in men is 0.054 L/s per kilogram, which is 3.2 L/min per kilogram, so ten kilograms moves that one prediction by 32 L/min while the other three do not move at all. That disagreement about whether weight belongs in a peak flow equation is itself informative: it is one of the reasons four equations fitted on four populations cannot be reconciled into one.

What is a lower limit of normal, and why does only one equation have one?

A predicted value is the middle of a reference population; a lower limit of normal is the fifth percentile of it, meaning that 5% of healthy subjects fall below. NHANES III publishes both, and the distance between them is the number that should sit beside every percentage of predicted: for a man of 40 at 170 cm its predicted value is 570 L/min and its fifth percentile is 443, so a blow 22.3% below predicted is still inside the reference range. The other three equations publish a predicted value and a residual error rather than a percentile limit. Nunn and Gregg’s normal limits reach the reader in absolute terms through the chart adapted from it: up to 100 L/min below predicted in men and 85 in women. The Madras paper publishes standard errors of 1.574 L/s in men and 1.013 L/s in women, which are 94 and 61 L/min. Every one of those numbers is wider than the gap between 100% and 80% of predicted.

Does a low peak flow mean I have asthma?

No. A low peak flow means a low peak flow. It can be narrowed airways from asthma, but it can equally be chronic obstructive lung disease, a chest infection, restriction from any cause, respiratory muscle weakness, obesity, pain, a blocked nose, or simply a poor blow — and peak flow is so effort-dependent that a weak or slow effort is indistinguishable from obstruction in the number alone. It also agrees poorly with spirometry: in a study at an Indian teaching hospital, peak flow percent predicted and forced expiratory volume percent predicted differed by more than five percentage points in about three quarters of patients with airflow obstruction, which means a peak flow is not a substitute for the test. Diagnosis needs a clinician, usually spirometry, and usually more than one occasion. This page does not diagnose anything and has no code path that could.

Why not just use one modern equation for everybody, like the GLI standard for spirometry?

Because no such thing exists for peak flow. The Global Lung Function Initiative’s 2012 reference equations, which did bring multi-ethnic reference values to spirometry, cover forced expiratory volume, forced vital capacity and their ratio and do not include peak expiratory flow. So peak flow is still being read against a patchwork of single-population equations from 1963 onwards, which is exactly the situation a 2004 review in the American Journal of Emergency Medicine described: across the published formulae, an 18-year-old man’s 100%-predicted value ranged from 501 to 730 L/min, a spread of 229 L/min, and a 35-year-old woman could be classified severe at 46% of predicted on one formula and moderate at 57% on another. That review recommended that recently published population-based equations should be the reference standard, which is the argument for an Indian reader using an Indian equation and for this page printing all four rather than choosing silently.

Can I use this for a child?

No. All four equations on this page are adult equations and the page refuses below 15 years. Childhood peak flow reference values are a separate problem: the usual childhood equations are height-only, there is a widely circulated arithmetic approximation whose primary source could not be established here, and the Indian paediatric literature has its own regional equations that are not the ones above. A child’s peak flow should be read against a paediatric reference and, better, against the child’s own personal best. Growth and weight questions for children on this site are the paediatric weight estimation page and the mid-parental height page.

Related calculators

References

Kodgule R, et al. Reference values for peak expiratory flow in Indian adult population using a European Union scale peak flow meter. J Postgrad Med 2014;60(2):123–9. doi:10.4103/0022-3859.132311. Multicentre study of 6,138 healthy Indian adults from varied geographic and socioeconomic backgrounds (3,720 men, 2,418 women), measured on an EU-scale meter. Derived equations, in litres per minute: men PEF = 3.206 × height(cm) − 1.807 × age; women PEF = 2.368 × height − 1.454 × age. Gender, age and height were the significant determinants. Validated on a fifth of the sample, 1,000 adults selected at random and stratified by height and age, with mean differences between predicted and measured of 1.85 L/min in men and 1.64 L/min in women. Source of the central comparison on this page: an Indian adult of average height and age was found to have approximately 30% lower PEF than the corresponding European adult using the Nunn and Gregg equation. CHECK PERFORMED: that statement is a percentage of the European value, and reproducing it in this page’s own engine at plausible average Indian adult dimensions gives 25.1% for a man of 35 at 165 cm, 26.4% for a man of 40 at 165 cm, 34.7% for a woman of 35 at 152 cm and 35.6% for a woman of 40 at 152 cm — which brackets the stated 30% and is the main evidence that the coefficients used here are the paper’s own. The page prints the gap against BOTH denominators, because 32.9% of the Indian figure and 24.8% of the European figure are the same 156 L/min. WHAT COULD NOT BE ESTABLISHED: the journal page could not be read from here, so the full author list, the exact age range of the sample and the residual standard deviation of the equations are not stated on this page. The study was first presented as Derivation of a predicted equation for peak expiratory flow (PEF) values in adult Indian population using EU scale peak flow meter (PFM) at the 2011 European Respiratory Society Annual Congress by R Kodgule, V Singh, B Saicharan, R Dhar, J Londhe, B Brashier, U Singh, U Hafiz, S Mukherjee, S Madas, S Salvi and P Koul; the journal’s own author order is not asserted here.
Nunn AJ, Gregg I. New regression equations for predicting peak expiratory flow in adults. BMJ 1989;298(6680):1068–70. PMID 2497892. A new model used to calculate regressions of PEF on age and height, from which predicted values were derived for men and women aged 15 to 85. The equations implemented here: men ln(PEF) = 0.544 ln(age) − 0.0151 age − 74.7/height(cm) + 5.48; women ln(PEF) = 0.376 ln(age) − 0.012 age − 58.8/height(cm) + 5.63, both giving litres per minute on the original Wright scale. This is the equation behind the normal-values chart printed on peak flow diaries across the United Kingdom. WHAT COULD NOT BE ESTABLISHED: the paper itself could not be read from here; the open archive copy was behind an automated challenge. The coefficients come from an independent clinical calculator citing the paper and were verified by their consequences rather than by sight — the maxima they imply, 36.0 years in men and 31.3 in women, reproduce the rise-then-fall shape the model exists to produce, and the height unit must be centimetres because metres collapses the exponential. One secondary source states the height unit as metres and is wrong.
Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med 1999;159:179–87. Read from the copy hosted by the United States Centers for Disease Control and Prevention (CDC Stacks 198173), which is why this equation is the one on this page whose coefficients were read first-hand. PEF equations take the form b0 + b1·age + b2·age² + b3·height². White men 20 and over: 1.0523 + 0.08272 age − 0.001301 age² + 0.00024962 height²; under 20: −0.5962 − 0.12357 age + 0.013135 age² + 0.00024962 height². White women 18 and over: 0.9267 + 0.06929 age − 0.001031 age² + 0.00018623 height²; under 18: −3.6181 + 0.60644 age − 0.016846 age² + 0.00018623 height². Coefficient of determination 0.7808 for white men and 0.5559 for white women. The fifth-percentile lower limit of normal uses the same equations with a reduced height² coefficient — 0.00017635 for men and 0.00012148 for women — which is the source of the lower-limit-of-normal row on this page. Equations are also published for African-American and Mexican-American subjects; no South Asian or Indian group is included, which is stated on the page rather than worked around.
Vijayan VK, Kuppurao KV, Venkatesan P, Sankaran K. Reference values and prediction equations for maximal expiratory flow rates in non-smoking normal subjects in Madras. Indian J Physiol Pharmacol 1993;37(4):291–7. 273 healthy non-smoking adults aged 15 to 63 resident in Madras, measured on a Transfer Test Model C spirometer with computer-aided analysis. Men (n=144, mean age 29.9 ± 12.7 years, height 166.1 ± 7.7 cm, weight 56.0 ± 12.7 kg): PEFR = −5.837 − 0.017 age + 0.064 height + 0.054 weight, multiple correlation 0.553, standard error 1.574 L/s. Women (n=129, mean age 30.1 ± 9.4, height 151.8 ± 5.8 cm, weight 45.7 ± 9.1 kg): PEFR = −0.179 − 0.006 age + 0.025 height + 0.021 weight, multiple correlation 0.275, standard error 1.013 L/s. All values in litres per second. The authors report that flow rates at lower lung volumes in men were similar to those reported for Caucasians while in women they were lower, and attribute part of the difference in women to indoor air pollution from childhood.
United States National Heart, Lung, and Blood Institute. Asthma Action Plan. NIH Publication No. 20-HL-5251, February 2021. The template a clinician completes with a patient, written in percentages of a personal best: green zone peak flow more than 80% of my best peak flow, yellow zone 50% to 79% of my best, red zone less than 50% of my best. Cited on this page for the convention and the fact that it is anchored to a personal best rather than to a predicted value, and for nothing else. The page prints 80% and 50% of a personal best as arithmetic, does not assign a zone and does not reproduce any action. Chosen as a United States Government work in preference to the equivalent material from other bodies, which is this project’s standing position.
United States National Library of Medicine, MedlinePlus patient instructions, Peak flow meter. Source of the method for establishing a personal best quoted on this page: take a peak flow each day for 2 to 3 weeks with asthma under control, between noon and 2 pm and after a quick-relief medicine, and use the highest. Also the source for technique: hold the breath, lips closed around the mouthpiece, tongue away from the hole, and blow out as hard and fast as possible in a single blow, the first burst of air being the part that counts.
Radeos MS, Camargo CA Jr. Predicted peak expiratory flow: differences across formulae in the literature. Am J Emerg Med 2004;22(7):516–21. The paper that establishes, in a Western context, the problem this page is built around. Across the published formulae it compared — Godfrey 1970, Polgar and Promhadat 1971, Gregg and Nunn 1973, Leiner 1963, Hankinson 1999 and Hsu 1979 — choosing different formulae gave an 18-year-old man a 100%-predicted PEF as low as 501 L/min and as high as 730 L/min, a difference of 229 L/min, and gave a 35-year-old woman a classification of severe (46% of predicted) on one formula and moderate (57%) on another. The authors recommended that recently published population-based equations should be the reference standard for asthma guidelines.
Miller MR, Dickinson SA, Hitchings DJ. The accuracy of portable peak flow meters. Thorax 1992;47:904–9 — the measurement work behind the 2004 scale change, cited on this page for the size and direction of the difference: the original Wright scale overestimates peak flow in the mid-range by up to 80 L/min and underestimates above about 650, so a blow reading 440 on a Wright-scale meter reads about 387 on an EU-scale one, and differences of up to 30% between the readings are described. Quoted here from the occupational asthma reference material that cites it; the paper itself was not read from here. The resulting standard is EN 13826, carried forward as ISO 23747, and meters sold in the United Kingdom and European Union since 2004 are marked to it.
Clement Clarke International, Peak expiratory flow rate — normal values, the nomogram adapted by the manufacturer for use with EU-scale (EN 13826 / ISO 23747) meters from Nunn AJ, Gregg I, BMJ 1989;298:1068–70. Not reproduced on this page in any form. Three statements printed on it are quoted, because each is a fact rather than expression: that in men readings up to 100 L/min lower than predicted are within normal limits and in women the equivalent figure is 85; that it covers ages 15 to 85, men 160 to 190 cm and women 152 to 183 cm; and that the values are derived from Caucasian populations. The manufacturer also distributes the Wright-to-EU scale conversion as a chart and an online converter, and this page does not reproduce, redraw, tabulate or curve-fit it.
Aggarwal AN, Gupta D, Jindal SK. The relationship between FEV1 and peak expiratory flow in patients with airways obstruction is poor. Chest 2006;130(5):1454–61. Cited on this page for one finding: peak flow percent predicted and forced expiratory volume in one second percent predicted were more than 5 percentage points apart in approximately three quarters of patients with airflow obstruction, in an Indian teaching hospital population. It is on the page because the percentage this page computes is routinely treated as interchangeable with a spirometric percentage, and in the same patients it is not.
European Respiratory Society, Global Lung Function Initiative. Cited on this page for a negative: data on PEF and other instantaneous flows were not collected for GLI-2012, so it is not possible for manufacturers to depict predicted contours for them. This is why peak flow has no modern multi-ethnic reference equation of the kind that exists for forced expiratory volume and forced vital capacity, and why this page computes four single-population equations rather than one equation with an ethnic coefficient.
LICENSING POSITION taken for this page, recorded because it determined what the page contains. The four regression equations are used freely: a fitted regression is a method and its coefficients are facts about data, which is the position this project has taken before and is what allows an equation to be used even where the paper carrying it is not open. A manufacturer’s nomogram and a manufacturer’s scale-conversion table are different — they are expression, they are commercial products, and no part of either is reproduced, redrawn, tabulated or approximated here; only the limits and ranges stated on the chart are quoted, because a stated limit is a fact. The zone convention and the personal-best method are taken from United States Government works (National Heart, Lung, and Blood Institute, and MedlinePlus from the National Library of Medicine) in preference to the equivalent material from other bodies, which is this site’s standing preference where a figure exists in more than one place. No guidance from the United Kingdom national institute is used: its open content licence is United Kingdom-only and forbids display of the licensed information next to advertising, and this site carries advertising. No World Health Organization material is used: its licence is NonCommercial and this site is commercial.
Derivations and checks performed for this page rather than taken from a source, recorded so they can be checked. (1) The European minus Indian gap, computed at 170 cm for men: 20 L/min (3.8%) at age 15, 74 (14.6%) at 20, 134 (27.4%) at 30, 156 (32.9%) at 40, a maximum absolute gap of 157.7 L/min at age 44.6, a maximum PERCENTAGE gap of 34.2% at age 48.8, then 155 (34.2%) at 50, 143 (32.7%) at 60 and 100 (25.1%) at 80. For women at 155 cm: 96 (27.7%) at 15, a maximum absolute gap of 163.7 L/min at age 42.2, a maximum percentage gap of 54.9% at age 54.0, and 129 (51.3%) at 80 — the absolute and percentage maxima do not coincide in either sex, because the Indian equation the percentage divides by is still falling. (2) The age of maximum predicted flow in Nunn and Gregg’s equation is 0.544/0.0151 = 36.0 years for men and 0.376/0.012 = 31.3 for women; the other three equations fall monotonically from 15. (3) The two NHANES III age branches meet at their switch points: 9.40042 against 9.40032 L/s at age 20 for men and 6.60720 against 6.60736 L/s at age 18 for women, which confirms the coefficients were transcribed correctly. (4) The two Indian equations cross at about 170 cm for a man of 40 at 65 kg: the EU-scale equation is 13.0 L/min above the Madras one at 150 cm, 0.4 above at 170 and 12.3 below at 190, so their near-identity at the page’s default inputs is a crossing and not agreement. They also separate with age, because their age slopes differ (1.807 against 1.02 L/min a year): at 170 cm the EU-scale figure is 20.0 L/min above the Madras one at 15 years and 31.1 L/min below it at 80. (5) For a man of 40 at 170 cm, 80% of the Indian predicted value falls at 378 L/min and 80% of Nunn and Gregg’s at 503, a 125 L/min difference in where the same convention sits; a blow of 500 is 105.8% of the Indian figure and 79.6% of the European one. (6) The NHANES III fifth percentile sits 22.3% below its own predicted value for a man of 40 at 170 cm and 23.8% below for a woman of 40 at 155 cm, both of which are wider than the gap between 100% and 80% of predicted.

CalcEngines health calculators are for education and for checking arithmetic that has already been decided elsewhere. They are not medical advice, they do not decide what to give, and they do not replace the judgement of a doctor, nurse, midwife or dietitian who knows the person in front of them. Every figure depends on the values you enter and on the assumptions stated on the page — check it against the prescription, the product label and your local policy before acting on it.