Building endurance

Put an exercise finding the way it tends to get retold: one minute of hard exercise does the work of forty-five minutes of steady exercise. That's this course's paraphrase, not a headline it collected. It comes from a real study, done well, and the study says a good deal less. Endurance training changes your heart and blood more than your muscles, and in healthy young adults intervals beat steady work by a little, not a lot. People who seem not to improve have often done too little or been measured too few times, though people really do differ in how much they gain. The last part of the lesson is how to start walking or running without talking yourself into an injury.

What fitness is, measured in oxygen

When exercise scientists say "fitness" in the endurance sense they usually mean one number: VO2max, the most oxygen your body can take in and use in a minute of all-out work. It's written in millilitres of oxygen per kilogram of body weight per minute (mL/kg/min), so a heavy person and a light one can be compared.

Lesson 2 gave you the MET, which the Compendium pegs to a standard resting uptake of 3.5 mL/kg/min.⁴ So a VO2max of 35 mL/kg/min is ten times that resting figure, or 10 METs. Keep that conversion in your pocket; it turns two gaps later in this lesson into something you can picture.

To use oxygen, your body has to move it along a chain. Air comes into the lungs. Oxygen crosses into the blood and rides on the red blood cells. The heart pumps that blood out, and how much it pumps a minute (the cardiac output) is the beats per minute times the amount pushed out on each beat (the stroke volume). The blood reaches the working muscle through tiny vessels, the capillaries, and inside each muscle cell, small structures called mitochondria use the oxygen to burn fuel and release energy. Any link in that chain could, in principle, be the one that sets the ceiling.

Hellsten and Nyberg's review names the same adaptation as the most important one, and gives its parts: "The functionally most important adaptation is the improvement in maximal cardiac output which is the result of an enlargement in cardiac dimension, improved contractility, and an increase in blood volume, allowing for greater filling of the ventricles and a consequent larger stroke volume."³ They add a change at the far end of the chain: the network of capillaries inside the muscle grows, which gives the muscle more surface and more time to take oxygen out of the passing blood.³

This course read all three reviews at abstract level only, so it can give you their conclusions but not the experiments under them.

The ceiling and what you do below it

The muscles do matter, just not mainly for the ceiling. Bassett and Howley are explicit: "Metabolic adaptations in skeletal muscle are, however, critical for improving submaximal endurance performance."¹ A long effort is done below the ceiling, at some fraction of it: "Running economy and fractional utilization of VO2max also affect endurance performance."¹ They single out the speed at the lactate threshold, which they call "the best physiological predictor of distance running performance."¹

Lactate needs a word here, because most people meet it as the waste that makes muscles burn. Brooks's review, read at abstract level, gives the current view: "Once thought to be a waste product of anaerobic metabolism, lactate is now known to form continuously under aerobic conditions."³² The lactate threshold is a point on a graded exercise test, found from blood samples, and a review of the talk test treats it as a near neighbour of lesson 2's ventilatory threshold: above either one, comfortable speech stops being likely.³⁰ Lesson 8 comes back to lactate and soreness.

So endurance has two parts, and it helps to keep them apart. The ceiling is set mostly by delivery: heart, blood volume, stroke volume. How close to the ceiling you can work for a long time leans heavily on the muscle (capillaries, mitochondria, the lactate threshold), and training moves both.

Fitness and survival: which sentence is this?

Lesson 1 separated three sentences that get spoken as one: exercise changes the body; active people live longer; becoming active will make you live longer. Fitness gives you a clean chance to use them.

A meta-analysis by Kodama and colleagues pooled 33 observational studies of healthy adults, about 103,000 people, with fitness measured as maximal capacity in METs. Its abstract, which is all this course read, reports that each 1-MET-higher level of fitness went with a relative risk of death of 0.87, about 13 per cent lower.⁵

A larger single study, Mandsager and colleagues', followed 122,007 patients sent to the Cleveland Clinic for a treadmill stress test, for a median of 8.4 years. The fittest group had an adjusted hazard ratio for death of 0.20 against the least fit.⁶ The authors' own limitation is the part worth reading twice: "The degree to which high CRF preselects patients with lower mortality vs causes a reduction in mortality is not discernible from our study."⁶ And these were people referred for a clinical test, not a sample of the public.

Steady work or intervals

There are two broad ways to train endurance. Continuous training, which the research literature calls moderate-intensity continuous training (MICT), means one sustained effort: a 45-minute ride, a 50-minute brisk walk. Interval training, HIIT for high-intensity interval training, means hard bouts broken by easier recovery.

The reason intervals exist is visible in the trial protocols themselves. You met Generation 100's two intensities in lesson 2. Its continuous sessions were about 50 minutes at about 70 per cent of peak heart rate; its interval sessions were four bouts of 4 minutes at about 90 per cent, with a 10-minute warm-up in front.¹⁰ Nobody asked those 70- to 77-year-olds to hold 90 per cent for 50 minutes. Breaking the hard work into pieces, with easier work between, is what lets a person spend a total of 16 minutes at an intensity they could not sustain in one go. (That's this course's reading of why the protocols are built the way they are; none of the sources read tested it directly.)³¹

Lesson 2's talk test marks the line: speech first becomes difficult at about the ventilatory threshold,²⁰ and the hard bouts are meant to go past it, which is again this course's reading of the protocols.³¹

Their conclusion, in full: "Endurance training and HIT both elicit large improvements in the VO2max of healthy, young to middle-aged adults, with the gains in VO2max being greater following HIT when compared with endurance training."⁷ That sentence and its numbers need three limits beside them. The people averaged 25 years old, with nobody over 45. The only outcome was VO2max. And the analysis used a statistical method, magnitude-based inference, that statisticians have since criticised. This course read neither that criticism nor more than the meta-analysis's abstract, so it flags the point and doesn't argue it.⁷

In people with heart and metabolic disease the gap looked larger. Weston, Wisløff and Coombes pooled 10 randomised trials of 273 patients with coronary artery disease, heart failure, high blood pressure, metabolic syndrome or obesity, in supervised programmes of 4 to 16 weeks.⁸ Their summary: "HIIT significantly improved CRF by almost double that of MICT (19.4% vs 10.3%)."⁸ (CRF is cardiorespiratory fitness.) In absolute terms the difference was about 3 mL/kg/min, close to a whole MET, against the third of a MET in Milanović's healthy young adults.⁸ And their own limitation, which belongs next to it every time that figure is repeated: "A limitation of this review is that 6 of the 10 studies reviewed are from the same research group. It is important that these findings are replicated in future studies at other institutions."⁸ Wisløff, one of the authors, is from that group.

One caution from its full text, which this course read. Most programmes were supervised throughout, three added a home programme, and "No studies described their adverse event monitoring protocol."⁸ The patients were included only if their disease was fairly well controlled or stable. The authors cite a cardiac rehabilitation study with one fatal cardiac event during moderate training and two non-fatal ones during high-intensity training: rare under either, and more frequent per hour at high intensity. They add that the study was far too small to settle the question, and still read the evidence as showing that "although there are limited data, HIIT appears to be well tolerated and safe."⁸ Their own condition is the part to keep: "if participants are adequately screened and the programme is prescribed and supervised by appropriately trained and qualified individuals, then it should be an achievable training option."⁸ They also ask for all-out sprint intervals, like the ones in the next section, to be kept apart from other interval training, because "there are concerns for the safety of this all-out approach in clinical populations."⁸ So the review doesn't show that hard intervals are safe for somebody with heart disease exercising alone, and its authors don't claim it does. For anybody in that position the callout at the top of this lesson applies, and the plan comes from their doctor or rehabilitation team.

The one-minute study, taken apart

Now the finding this lesson opened with. Gillen and colleagues, in a laboratory study, compared two programmes over twelve weeks.⁹ The protocols, in the paper's words: "The SIT protocol consisted of 3x20-second ‘all-out’ cycling efforts against 0.05kg/kg body mass, separated by 2 minutes of low-intensity cycling (50W). The MICT protocol consisted of 45 minutes of continuous cycling at ~70% HRmax."⁹ SIT is sprint interval training. With warm-up and cool-down the sessions took 10 minutes and 50 minutes, three times a week.⁹ The resistance was scaled to body weight, and the abstract puts each sprint at about 500 watts of power on the bike, against 50 for the easy pedalling between sprints and about 110 for the steady ride. That's what "all-out" means here.⁹

Three 20-second sprints is one minute of hard work a session, and the authors framed it that way themselves: "SIT involved 1 minute of intense exercise within a 10-minute time commitment, whereas MICT involved 50 minutes of continuous exercise per session."⁹ Their summary sentence: "we report that a SIT protocol involving 3 minutes of intense intermittent exercise per week, within a total time commitment of 30 minutes, is as effective as 150 minutes per week of moderate-intensity continuous training for increasing insulin sensitivity, cardiorespiratory fitness and skeletal muscle mitochondrial content in previously inactive men."⁹

Take it apart with three questions. Lesson 4 turns them into this course's standing question for every training claim; here they're just tools.

Trained whom? Twenty-seven sedentary men were enrolled and assigned to a group; the paper says "assigned" and does not describe randomisation. One man dropped out of each training group, which left 9 in the sprint group, 10 in the continuous group and 6 controls. They were 27 years old on average, give or take 8.⁹ No women.

For how long? Twelve weeks, three supervised sessions a week, on a laboratory bike.⁹

The study is careful, and the one-minute framing is in its own abstract. Retold as one minute equalling forty-five, it loses the men, the laboratory, the twelve weeks and the word "all-out", and a finding of no detectable difference in 9 and 10 men turns into a claim that the two are equal.

Those were young men, supervised in a laboratory, and an all-out effort isn't where a beginner starts. It's the kind of session Weston's authors flagged above. Lesson 8 comes back to a case series of people whose first hard spin class put them in hospital, which is what the last line of the callout at the top of this lesson is about.

Will people do it? Two researchers, one debate

Whether intervals work isn't really the argument any more. Biddle and Batterham wrote a formal debate paper in 2015, each writing his own side, and they agreed on that much in the abstract: "The efficacy of high-intensity interval training for a broad spectrum of cardio-metabolic health outcomes is not in question. Rather, the effectiveness of this form of exercise is at stake."¹¹ Efficacy means it works when people do it as prescribed; effectiveness means it works when you recommend it to a population who then live their lives.

Biddle, against: "There is currently no evidence supporting HIT as a viable public health strategy."¹¹ His reasons: the studies are efficacy-only, short and small, and few are randomised trials.¹¹ And on how exercise feels: "we need to boost people’s positive feelings about exercise. Making it harder and more painful is unlikely to do this."¹¹

Biddle's worry rests on dual mode theory, a model of how exercise feels. In his summary of it, feelings during hard exercise "unify into a negative trend as the intensity of exercise approaches each individual’s functional limits".¹¹ In plain terms, the closer you get to your limit, the worse it feels, for nearly everybody. He cites a trial of inactive overweight adults in which "adherence was particularly low in the two high intensity conditions."¹¹

Batterham, for, answered that point: "The bottom line is that dual mode theory applies to continuous exercise above the ventilatory threshold, not to low-volume HIT with its built-in recovery periods."¹¹ His reason: "Anticipation of impending recovery, and the recovery period per se, results in more positive affect than continuous vigorous intensity exercise."¹¹ Affect is how you feel. He cites a small study, by Jung and colleagues, in which the intervals were harder than a continuous vigorous session and still felt better. Biddle reads the same study the other way: there, "HIT showed progressively negative affective reactions compared to continuous moderate physical activity".¹¹ So the two read the same small study in opposite directions. Batterham argued too that for an unfit person, brisk walking uphill can count as high intensity, because a modest absolute effort is a large share of a low capacity.¹¹ His conclusion: "In summary, practical, enjoyable and scalable HIT protocols exist that could form a viable population health strategy, implemented within an evolutionary health promotion framework."¹¹

The longest real-world test sits between them. In Generation 100, over five years, "Adherence to prescribed exercise after one, three, and five years was, respectively, 201 (50%), 196 (49%), and 187 (47%) of participants in the HIIT group and 244 (63%), 213 (55%), and 197 (51%) in the MICT group."¹⁰ About half of the interval group kept to its prescription for five years, and so did about half of the continuous group, which led by 13 points at one year and by 4 at five. That sits between the two cases. The 70- to 77-year-olds in that trial were volunteers, mostly healthy, and supervised or guided.¹⁰

So sort it. That interval training raises fitness in trials, as much as continuous training or a little more, in less time, is established. Whether the public will do it is a contested empirical question with evidence on both sides. And whether health services should promote it is partly a value question about priorities, which this course does not settle.

Why some people seem not to improve

Almost everyone who trains for endurance knows somebody who swears it does nothing for them. The idea has a serious scientific history, and three different explanations, and they have very different consequences for what you'd do next.

The family study

The HERITAGE Family Study put 481 sedentary white adults from 98 two-generation families through 20 weeks of the same cycle training, and tested VO2max twice before and twice after.¹² From the abstract, which is all this course read: "The mean increase in VO(2max) reached approximately 400 ml/min, but there was considerable heterogeneity in responsiveness, with some individuals experiencing little or no gain, whereas others gained >1.0 l/min."¹² Gains ran in families, and the authors reported "a maximal heritability estimate of 47% for the VO(2max) response".¹² Their conclusion: "We conclude that the trainability of VO(2max) is highly familial and includes a significant genetic component."¹² Two details to hold. "Maximal" means 47 per cent is the upper bound their model allowed, not a point estimate. And the study had no group that didn't train.

The dose studies

Two later studies asked a different question: did the so-called non-responders get enough?

Montero and Lundby recruited 78 untrained young men, aged 18 to 35, into five groups doing one to five 60-minute sessions a week of supervised cycling for 6 weeks.¹⁴ The men chose their own group, so the groups weren't randomised, and there was no group that didn't train. A man counted as a non-responder if his maximal cycling power changed by no more than the test's typical error, about 4 per cent either way. The results, from the abstract: "In groups 1, 2, 3, 4 and 5, 69%, 40%, 29%, 0% and 0% of individuals, respectively, were non-responders. After the second ET period, non-response was eliminated in all individuals."¹⁴ The second period was another 6 weeks with two extra sessions a week, for the people who had not responded the first time.

The share of men whose maximal cycling power didn't change by more than the test's typical error after six weeks, by weekly dose. Drawn from the figures in Montero and Lundby's paper;¹⁴ the groups held 14 to 17 young men each, so the bars are 11, 6, 4, 0 and 0 people. The outcome is cycling power, not VO2max.

Read the chart from left to right. At an hour a week, more than two thirds of these men didn't measurably improve in six weeks. At four hours, none failed to. The paper's key points add that "The magnitude of CRF improvement is primarily attributed to changes in haemoglobin mass."¹⁴ Haemoglobin is the oxygen-carrying part of the red blood cells, so that ties straight back to the blood you met at the start of this lesson. The authors' own limitation sets the scope: "Findings were obtained from a sample of healthy young males."¹⁴

Ross, de Lannoy and Stotz, in a trial this course read at abstract level, separated amount from intensity over 24 weeks, in 121 sedentary, middle-aged, abdominally obese adults.¹³ Non-responders were 38.5 per cent with a low amount at low intensity, 17.6 per cent with a high amount at low intensity, and none with a high amount at high intensity. In their words: "At a fixed amount of exercise, increasing the exercise intensity eliminated nonresponse (P=.001)."¹³ They analysed only people who completed at least 90 per cent of their sessions, so the study tells you what happens if you do the training, not what happens if you're told to.

The measurement problem

The third explanation is statistical, and it applies even when nobody's genes or dose is at fault. Every test of fitness varies from day to day, even with the best equipment. Atkinson and Batterham put the consequence bluntly: "We highlight the impact of within-subject random variation, which is inevitable even with 'gold-standard' measurement tools/protocols and sometimes so substantial that it explains all apparent individual response differences."¹⁵ And they say what you'd need to measure true differences between people: "True individual response differences are quantified only by comparing the SDs of changes between intervention and comparator arms."¹⁵ In plain terms, you need a group who did not train, to see how much people's scores wander on their own.

To see why, take a made-up case, which is this course's illustration rather than anyone's data.³¹ Suppose a man's true cycling power rises by 7 per cent over six weeks, comfortably more than the line of about 4 per cent that Montero and Lundby drew. He tests 3 per cent high on a good day at the start and 1 per cent low on an ordinary day at the end. His measured change is then about 3 per cent (107 × 0.99 ÷ 103 is 1.028), under the line. He'd be counted a non-responder, and he did respond.

Williamson, Atkinson and Batterham then went looking for that comparison in the family study: "We located over 180 publications that resulted from the HERITAGE Family Study, but we could not find a comparator arm in any of these studies."¹⁶ In one older study they could re-analyse, which did have a comparison group, "The standard deviation of change was, in fact, larger (±5.6 mL/kg/min) for the comparator than the intervention group (±3.7 mL/kg/min)."¹⁶ The people who didn't train varied more than the people who did. Their conclusion: "true inter-individual differences in response cannot be quantified, let alone appraised for clinical relevance."¹⁶ This course read the Atkinson and Batterham paper in full and the Williamson review at abstract level.

The same test cuts the other way too. Montero and Lundby had no untrained comparison group either, and classed each man from a single test after training (against two at the start), so their individual responders and non-responders are counted in just the way Atkinson and Batterham warn about. What their design adds is the second period, in which the same men trained more and all of them then improved. And this course read no reply from the HERITAGE investigators to the statistical critique, so the family study's side of that argument isn't given here.

Sorting the three

These three views don't all contradict each other, and it helps to see exactly where they meet. That people's gains differ is established; every study here found it. That some people can't improve at all, given enough training, is contested, and on the sources read it's weakly supported: in both dose studies, non-response vanished when the dose went up, though their individual counts rest on the kind of measurement the statisticians question. Lundby and colleagues put the dose view in its strongest form: "All previously untrained individuals will respond to endurance exercise training", in VO2max, provided the stimulus exceeds a certain volume or intensity.² That's the position of the researchers who hold it, stated as theirs. How much of the variation between people is genetic is also contested: the family study says a lot, and the statisticians say that without a comparison group it couldn't measure how much people truly differ, which is the thing a heritability estimate divides up.

Zone 2: a claim, and only one side of it read

If you've listened to fitness podcasts in the last few years you'll have met zone 2: low-intensity work below the lactate threshold, which popular advice has promoted as the best way to build the muscle's mitochondrial and fat-burning capacity. That description comes from the critics' review below,¹⁹ and the reason is the first thing to know about this section. This course read no statement by a proponent of zone 2 training setting out their case in their own words, and no training trial in which a proponent reports zone 2 beating other intensities. The popular advocates, in books and podcasts, were not read. So this section can't give you the case for zone 2 at full strength, and it won't improvise one.¹⁷

What it did read is three things.

The physiology the advocacy draws on. San-Millán and Brooks tested 22 professional endurance cyclists, 20 moderately active men and 10 men with metabolic syndrome, once each on a cycling test: a snapshot comparison, not a training study. Higher blood lactate went with less fat burned and more carbohydrate, across all three groups, and they proposed lactate and fat-burning measurements as an indirect way to assess metabolic flexibility, which they describe as the ability to burn both fats and carbohydrates.¹⁷ The phrase "zone 2" does not appear anywhere in the paper; this course searched the full text for it.¹⁷

A definition, from people sympathetic to the idea. Sitko and colleagues, fourteen sport scientists and professional endurance coaches, wrote a short commentary, which this course read at abstract level. They note that "despite the popularity of this concept, there is no clear consensus among coaches, athletes, and scientists regarding the definition of zone 2 training."¹⁸ Their working definition: "zone 2 training should preferably be performed at intensities located immediately below the first lactate or ventilatory threshold".¹⁸ And their own hedge: "These expected adaptations might not be unique to zone 2 and could also be induced with sessions performed at slightly higher and lower intensities."¹⁸

A critique. Storoschuk, Moran-MacDonald, Gibala and Gurd reviewed whether zone 2 is the best way for ordinary people to improve mitochondria and fitness. On where the recommendation comes from: "These recommendations largely stem from observational data of elite endurance athletes who engage in large volumes of Zone 2 training and possess high mitochondrial and fatty acid oxidative capacity."¹⁹ Their conclusion, in full: "We conclude that current evidence does not support Zone 2 training as the optimal intensity for improving mitochondrial or fatty acid oxidative capacity. Further, evidence suggests prioritizing higher exercise intensities (> Zone 2) is critical to maximize cardiometabolic health benefits, particularly in the context of lower training volumes."¹⁹ Count the side that's speaking. The second sentence is a claim of its own, for harder work, and Gibala is one of the leading interval-training researchers and the last author of the one-minute study above. The review also declares that he advises, and holds equity in, a company whose services partly relate to exercise.¹⁹ So this is one camp's reading of the evidence, from a review this course read at abstract level. And read the scope: "optimal", and "the general population" in the title. The review doesn't say zone 2 does nothing.

One more thing not to mix up with zone 2: the "fat-burning zone" on a gym machine's display. In 761 untrained adults aged 41 to 68, each tested once on a bike, Pühringer and Ring-Dimitriou found the intensity of greatest fat burning lower than the aerobic threshold, at 35 to 43 per cent of peak work rate against 47 to 51, and the two agreed only weakly from person to person. This course read the abstract.³³

There's one practical bridge. Sitko's group puts zone 2 just below the first lactate or ventilatory threshold,¹⁸ and a review of the talk test says "comfortable speech is likely possible (equivocal or last positive talk test stage) when exercise intensity is below the ventilatory or lactate threshold".³⁰ Put the two together and zone 2 is roughly the top of the range where you can still talk comfortably. That last step is this course's, and it assumes the review's threshold is the first of the two. Both boundaries are fuzzy at the edges.³¹

So the classification is: contested, with only the critics' side read in its own words. What would settle it is a training trial in ordinary adults that gives one group zone 2 work and another group harder work for the same time, for months, and measures fitness and mitochondrial capacity. None was read for this course. If you enjoy steady, conversational training, nothing here says it does you no good: Milanović's meta-analysis found continuous training raised VO2max a lot in healthy adults under 45, though this course can't tell from the abstract whether its trials ran at a talking pace.⁷ What's unsupported, on what was read, is the word "best".

Starting to walk or run

Everything so far is about what adapts. The rest is about starting: a first month without quitting or getting hurt.

Steps, and where 10,000 came from

Walking is the easiest start, and the most counted. Lee and colleagues, in a full-text paper this course read, are careful about its origin: "However, the origin of the goal of 10 000 steps per day is unclear. It likely derives from the trade name of a pedometer sold in 1965 by Yamasa Clock and Instrument Company in Japan called Manpo-kei, which translates to “10 000 steps meter” in Japanese."²² Note "is unclear" and "likely". A later paper from the same research network states it more flatly, that the number "originates from a marketing campaign in Japan".²¹ This course did not read any 1965 advertisement or company record, so "likely" is the honest word.

When researchers counted, Paluch and colleagues pooled 15 cohorts, 47,471 adults with steps measured by a device, followed for a median of 7.1 years.²¹ Risk of death fell "among adults aged 60 years and older with increasing number of steps per day until 6000–8000 steps per day and among adults younger than 60 years until 8000–10 000 steps per day."²¹ In Lee's study of 16,741 women averaging 72, the curve levelled at about 7,500.²² Both are observational, and Paluch's own limitation applies to both: "The data are derived from observational studies; therefore, causal inferences cannot be made."²¹ That's lesson 1's second sentence again, not its third.

So a probably commercial round number sits in about the right region for younger adults and above the levelling point for older ones, in observational data, and nothing in those curves makes 10,000 a line to cross.

Couch to 5K, and what it's built on

The NHS's Couch to 5K is a free beginner running plan. It runs 9 weeks, three runs a week with a rest day between, each run bracketed by a 5-minute walk to warm up and cool down.²³ Week 1: "Run for 1 minute and walk for 1 minute 30 seconds. Do this 7 times in total", then a final minute of running.²³ By week 5 there's a continuous 20-minute run; by week 9, three 30-minute runs a week.²³

Notice what the plan's end point is. It isn't a distance: "By the end of the 9 weeks, you’ll be running for 30 minutes."²³ Plenty of beginners won't cover 5 km in 30 minutes, and the plan does not need them to. It's built on time. It's also built on patience: "Remember, it’s not about speed, it’s about building stamina." And "if you need to repeat any of the runs or weeks, that’s totally okay."²³

What rests under it? Relph and colleagues, studying a modified version of the programme in a paper this course read at abstract and introduction level, put it directly: "Interestingly, there is no empirical evidence to support the design of Couch-to-5k".²⁴ That does not mean the design is bad; it means the weeks, minutes and ratios are a sensible convention, not a tested prescription. Of their 110 runners, mostly women with an average age of 47, "Twenty-one injuries were reported during the programme (19%). Previous injury increased the risk of new injury (OR 7.56 95% CI from 2.06 to 27.75). Only 27.3% completed the programme."²⁴ That odds ratio's interval is very wide, and 64 per cent of the group already had some running experience.

How often beginners get hurt

It's better to know the injury numbers before you start than to meet them by surprise. Videbæk and colleagues pooled 13 studies: "The weighted estimates revealed novice runners faced a significantly greater injury rate of 17.8 (95 % CI 16.7–19.1) than recreational runners, who sustained 7.7 (95 % CI 6.9–8.7) running-related injuries per 1000 h of running."²⁵ They suggest why the novice figure is higher: "If novice runners exceed 8–13 weeks without injury, they may well have adapted to running and face a lower injury risk after this period, even though they may spend more time running."²⁵ And they warn that the studies defined injury differently, from one lost day of running to a week.²⁵

A rate per 1,000 hours is hard to feel, so turn it into a programme. This is this course's arithmetic, not the paper's.³¹ Add up Couch to 5K's running minutes, walks left out: 8 a run in week 1, rising to 30 in week 9, three runs a week, comes to 489 minutes over the nine weeks, about 8.2 hours.²³ At 17.8 injuries per 1,000 hours, that's 8.2 × 17.8 ÷ 1,000, about 0.15 injuries per runner, or roughly one beginner in seven. The two beginner studies in this lesson saw more: 21 injuries among Relph's 110 runners, and a fifth of runners injured in the trial below, whose own rates, 30 and 38 per 1,000 hours, ran well above the pooled figure. Pooled rates also mix studies that counted an injury after one lost day with studies that waited a week. So treat one in seven as a low estimate for a first programme, not a forecast.

What GRONORUN did and did not show

Start with what was compared. Not a careful plan against a reckless one, but a 13-week build-up against an 8-week one, in 532 novices, 57.5 per cent of them women, preparing for the same 4-mile event.²⁶ Both groups ran three times a week. Both were told to run "only at a comfortable pace at which they could converse without breathlessness".²⁶ That's lesson 2's talk test, given to both groups as an instruction.

Then the numbers. As a share of runners, the two groups were almost identical. Per 1,000 hours of running, the graded group's rate was 30 and the standard group's 38, but the ranges around those figures overlapped widely (22 to 38, and 27 to 49) and the difference was not statistically significant.²⁶ One other difference: more of the standard group never started at all, 32 of 268 against 14 of 264.²⁶

Then the authors' own explanation of the null result, from the full text this course read. The contrast between the two programmes "may have been too small to cause an effect", and nobody measured how hard people actually ran.²⁶

So what does it show? That stretching a beginner's build-up from 8 weeks to 13 under the 10 per cent rule did not detectably reduce injuries, in these runners, preparing for this distance, when both groups ran at a conversational pace. What it does not show is that pacing does not matter. It did not test a reckless programme at all. A Danish cohort of 874 novices with GPS watches, read at abstract level, offers the nearest thing to that: for "distance-related" injuries, runners who increased by more than 30 per cent had a hazard ratio of 1.59 against those under 10 per cent, with an interval from 0.96 to 2.66, which isn't statistically significant.²⁷ Its authors call the study exploratory and suggest that "novice runners may be well advised to progress their weekly distances by less than 30% per week over a 2-week period."²⁷ And a systematic review of 31 studies, also read at abstract level, concluded, "It was not possible to identify which training errors were related to running related injuries."²⁸

Put together: nothing read shows 10 per cent is a special number. Very large jumps may matter for some injuries, weakly. The one risk factor that stood out in the newest beginner study was a previous injury.²⁴ This course's reading of all that, for a beginner, is dull: run at a pace you can talk at, build gradually by any sensible amount, repeat a week when it felt hard, and treat an old injury as a reason for extra patience.

Shoes and pronation

One more shop-floor claim. You may have been told your foot rolls inwards too much (overpronates) and that you need a special "stability" shoe. Nielsen and colleagues gave 927 Danish novices the same neutral shoe, measured each foot's posture, and tracked a year of running by GPS; 252 were injured.²⁹ Their conclusion, from the abstract: "The results of the present study contradict the widespread belief that moderate foot pronation is associated with an increased risk of injury among novice runners taking up running in a neutral running shoe. More work is needed to ascertain if highly pronated feet face a higher risk of injury than neutral feet."²⁹ Keep the second sentence. Very few feet in the study were highly pronated, so the question for them is open.

Six things people get wrong about endurance

"You need an hour of cardio to get fitter." Gillen's sprint group raised VO2peak as much as the continuous group in 10-minute sessions,⁹ in sedentary young men in a laboratory, so mind the scope. What the dose study does say is that the weekly total matters: at one 60-minute session a week, most of Montero and Lundby's young men didn't measurably improve in six weeks.¹⁴

"HIIT is simply better." In healthy young adults it added about a third of a MET over continuous training, and both raised VO2max a lot.⁷ Whether people stick to it is the live argument.¹¹

"Some people just don't respond." Non-response disappeared in both dose studies when the dose went up,¹³¹⁴ and much apparent non-response may be measurement noise,¹⁵ a point that applies to the dose studies' own individual counts as much as to the family study's. Differences between people are real and may run in families; a complete inability to improve is, on these sources, weakly supported.

"Never increase more than 10 per cent a week." The one randomised test found no difference.²⁶ The rule is unproven, not disproven.

"Overpronators need special shoes." Not for moderate pronation, in one large cohort of beginners.²⁹

"You need 10,000 steps." A number of uncertain, probably commercial, origin, which the mortality curves put about right for younger adults and above the levelling point for older ones.²¹²²

Practice

Connections

Back. Lesson 1 gave you the three sentences, and this lesson placed fitness and survival in the second of them. Its Generation 100 trial came back here for its interval arm and its adherence figures. Lesson 2 gave you the MET, which turned oxygen uptake into something you can compare, and the talk test, which came back here as a rough marker for zone 2 and as the pacing instruction in the GRONORUN trial. Logic and Argument lesson 5 taught you to judge a generalisation by its sample; the one-minute study is a sample of nine men. Mental Fitness lesson 2 asked what the comparison group got, and the non-responder debate asks whether there was one. Nutrition lesson 4 made the point about diets that the dose studies make about fitness: an average difference between groups says little about any one person in them.

Forward. Lesson 4 asks the same questions of strength. Muscles adapt to lifting differently from the way the heart and blood adapt to endurance work, and the three questions you used on the one-minute study become this course's standing question there, because the strength literature's defaults are narrower still.

Go deeper

• Biddle and Batterham, "High-intensity interval exercise training for public health: a big HIT or shall we HIT it on the head?", 2015, free. This course read it in full. It's the rare paper where both sides of a live argument write their own case in the same document, and it's short enough for an evening.
• Gillen and colleagues, "Twelve weeks of sprint interval training...", 2016, free in PLoS One. Read in full here. Read the methods section yourself and see how much of what you've heard about "one minute of exercise" survives it.
• Videbæk and colleagues, "Incidence of running-related injuries per 1000 h of running in different types of runners", 2015, free through PubMed Central. Read in full here. The clearest single source on how often beginners get hurt and why the numbers are hard to compare.
• NHS, Couch to 5K. Both pages read in full here. If you want to run, it's a free, plain, time-based plan, and you now know what it rests on.
• W. L. Kenney, J. H. Wilmore and D. L. Costill, Physiology of Sport and Exercise, 9th edition, Human Kinetics, 2025. The standard textbook, and its chapter on adaptations to aerobic and anaerobic training covers this lesson's physiology at length. This course saw only its table of contents, so it can point you to the book but can't vouch for any page of it.

Sources

1. D. R. Bassett and E. T. Howley, "Limiting factors for maximum oxygen uptake and determinants of endurance performance", Med Sci Sports Exerc 32(1), 2000, doi 10.1097/00005768-200001000-00012. Read level: abstract only. Supports: delivery as the primary limit, muscle adaptations and submaximal performance, lactate-threshold speed.
2. C. Lundby, D. Montero and M. Joyner, "Biology of VO2max: looking under the physiology lamp", Acta Physiol 220(2), 2017, doi 10.1111/apha.12827. Read level: abstract only. Supports: red cell volume and stroke volume; the authors' position that all untrained people respond given enough stimulus.
3. Y. Hellsten and M. Nyberg, "Cardiovascular adaptations to exercise training", Compr Physiol 6(1), 2015, doi 10.1002/cphy.c140080. Read level: abstract only. Supports: maximal cardiac output and its parts; capillary growth.
4. S. D. Herrmann, E. A. Willis, B. E. Ainsworth and others, "2024 Adult Compendium of Physical Activities", J Sport Health Sci 13(1), 2024, doi 10.1016/j.jshs.2023.10.010. Read level: full text. Supports: the 3.5 mL/kg/min standard.
5. S. Kodama and others, "Cardiorespiratory fitness as a quantitative predictor of all-cause mortality and cardiovascular events in healthy men and women: a meta-analysis", JAMA 301(19), 2009, doi 10.1001/jama.2009.681. Read level: abstract only. Supports: 33 studies, about 103,000 people, relative risk 0.87 per MET.
6. K. Mandsager and others, "Association of Cardiorespiratory Fitness With Long-term Mortality Among Adults Undergoing Exercise Treadmill Testing", JAMA Netw Open 1(6), 2018, doi 10.1001/jamanetworkopen.2018.3605. Read level: full text; abstract, methods, results summary, limitations. Supports: the population, follow-up, hazard ratio and the preselection limitation.
7. Z. Milanović, G. Sporiš and M. Weston, "Effectiveness of high-intensity interval training (HIT) and continuous endurance training for VO2max improvements", Sports Med 45(10), 2015, doi 10.1007/s40279-015-0365-0. Read level: abstract only. Supports: the trials, ages, gains and conclusion. The criticism of its statistical method was not read, and the body says so.
8. K. S. Weston, U. Wisløff and J. S. Coombes, "High-intensity interval training in patients with lifestyle-induced cardiometabolic disease", Br J Sports Med 48(16), 2014, doi 10.1136/bjsports-2013-092576. Read level: full text. Supports: the trials, patients, supervision and home programmes, figures, limitations, the cardiac rehabilitation events and rates, which it cites from a study this course knows only through Weston, and the authors' own reading of safety, their condition and their concern about all-out sprints.
9. J. B. Gillen, B. J. Martin, M. J. MacInnis, L. E. Skelly, M. A. Tarnopolsky and M. J. Gibala, "Twelve weeks of sprint interval training improves indices of cardiometabolic health similar to traditional endurance training despite a five-fold lower exercise volume and time commitment", PLoS One 11(4), 2016, doi 10.1371/journal.pone.0154075. Read level: full text. Supports: everything said about the one-minute study, including the authors' own one-minute framing and the power figures from the abstract.
10. D. Stensvold and others, "Effect of exercise training for five years on all cause mortality in older adults, the Generation 100 study", BMJ 371, 2020, m3485, doi 10.1136/bmj.m3485. Read level: full text. Supports: the prescriptions, ages and adherence.
11. S. J. H. Biddle and A. M. Batterham, "High-intensity interval exercise training for public health: a big HIT or shall we HIT it on the head?", Int J Behav Nutr Phys Act 12, 2015, 95, doi 10.1186/s12966-015-0254-9. Read level: full text. Supports: the common ground, each side's position, Biddle's summary of dual mode theory, the trial he cites, Batterham's reason and his reading of Jung and colleagues' study, and Biddle's reading of the same study. Jung's study itself was not read.
12. C. Bouchard, P. An, T. Rice and others, "Familial aggregation of VO2max response to exercise training: results from the HERITAGE Family Study", J Appl Physiol 87(3), 1999, doi 10.1152/jappl.1999.87.3.1003. Read level: abstract only.
13. R. Ross, L. de Lannoy and P. J. Stotz, "Separate effects of intensity and amount of exercise on interindividual cardiorespiratory fitness response", Mayo Clin Proc 90(11), 2015, doi 10.1016/j.mayocp.2015.07.024. Read level: abstract only.
14. D. Montero and C. Lundby, "Refuting the myth of non-response to exercise training", J Physiol 595(11), 2017, doi 10.1113/JP273480. Read level: full text, free on PubMed Central (PMC5451738), read for the abstract, key points, methods, the Figure 1 legend and the limitations. Supports: the participants (78 untrained young men aged 18 to 35), the self-chosen groups and the absence of an untrained group, the ±3.96 per cent typical error, the percentages and group sizes in the chart, haemoglobin mass, and the authors' limitation. The article carries a later correction notice, which this course did not read.
15. G. Atkinson and A. M. Batterham, "True and false interindividual differences in the physiological response to an intervention", Exp Physiol 100(6), 2015, doi 10.1113/EP085070. Read level: full text and abstract.
16. P. J. Williamson, G. Atkinson and A. M. Batterham, "Inter-individual responses of maximal oxygen uptake to exercise training: a critical review", Sports Med 47(8), 2017, doi 10.1007/s40279-017-0680-8. Read level: abstract only.
17. I. San-Millán and G. A. Brooks, "Assessment of metabolic flexibility by means of measuring blood lactate, fat, and carbohydrate oxidation responses to exercise in professional endurance athletes and less-fit individuals", Sports Med 48(2), 2018, doi 10.1007/s40279-017-0751-x. Read level: full text (the author manuscript). This course read no proponent's statement of the zone 2 case and no proponent training trial, and deliberately didn't read the popular advocates.
18. S. Sitko and others, "What is 'zone 2 training'?: experts' viewpoint on definition, training methods, and expected adaptations", Int J Sports Physiol Perform 20(11), 2025, doi 10.1123/ijspp.2024-0303. Read level: abstract only.
19. K. L. Storoschuk, A. Moran-MacDonald, M. J. Gibala and B. J. Gurd, "Much ado about zone 2", Sports Med 55(7), 2025, doi 10.1007/s40279-025-02261-y. Read level: abstract only. Supports: the popular description of zone 2 it tests, the two-sentence conclusion, and Gibala's declared interest, which is in the PubMed record's declarations.
20. R. Persinger, C. Foster, M. Gibson, D. C. Fater and J. P. Porcari, "Consistency of the talk test for exercise prescription", Med Sci Sports Exerc 36(9), 2004, PMID 15354048. Read level: abstract only.
21. A. E. Paluch, S. Bajpai, D. R. Bassett and others, "Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts", Lancet Public Health 7(3), 2022, doi 10.1016/S2468-2667(21)00302-9. Read level: full text; abstract, introduction, age results, discussion, limitations.
22. I-M. Lee, E. J. Shiroma, M. Kamada, D. R. Bassett, C. E. Matthews and J. E. Buring, "Association of Step Volume and Intensity With All-Cause Mortality in Older Women", JAMA Intern Med 179(8), 2019, doi 10.1001/jamainternmed.2019.0899. Read level: full text; abstract, introduction, limitations, conclusion. Lee's own references for the pedometer, and any 1965 primary source, were not read.
23. NHS, "Get running with Couch to 5K" and "Couch to 5K running plan", nhs.uk, accessed 24 September 2026. Read level: both full pages.
24. N. Relph, S. L. Taylor, D. L. Christian, P. Dey and M. B. Owen, "Couch-to-5k or Couch to Ouch to Couch!?", Int J Environ Res Public Health 20(17), 2023, 6682, doi 10.3390/ijerph20176682. Read level: abstract and introduction; results skimmed only for the figures also in the abstract.
25. S. Videbæk, A. M. Bueno, R. O. Nielsen and S. Rasmussen, "Incidence of running-related injuries per 1000 h of running in different types of runners", Sports Med 45(7), 2015, doi 10.1007/s40279-015-0333-8. Read level: full text.
26. I. Buist, S. W. Bredeweg, W. van Mechelen, K. A. Lemmink, G. J. Pepping and R. L. Diercks, "No effect of a graded training program on the number of running-related injuries in novice runners: a randomized controlled trial", Am J Sports Med 36(1), 2008, doi 10.1177/0363546507307505. Read level: full text.
27. R. O. Nielsen, E. T. Parner, E. A. Nohr, H. Sørensen, M. Lind and S. Rasmussen, "Excessive progression in weekly running distance and risk of running-related injuries", J Orthop Sports Phys Ther 44(10), 2014, doi 10.2519/jospt.2014.5164. Read level: abstract only.
28. R. O. Nielsen, I. Buist, H. Sørensen, M. Lind and S. Rasmussen, "Training errors and running related injuries: a systematic review", Int J Sports Phys Ther 7(1), 2012, PMC3290924. Read level: abstract only.
29. R. O. Nielsen, I. Buist, E. T. Parner and others, "Foot pronation is not associated with increased injury risk in novice runners wearing a neutral shoe", Br J Sports Med 48(6), 2014, doi 10.1136/bjsports-2013-092202. Read level: abstract only.
30. J. L. Reed and A. L. Pipe, "The talk test: a useful tool for prescribing and monitoring exercise intensity", Curr Opin Cardiol 29(5), 2014, doi 10.1097/HCO.0000000000000097. Read level: abstract only. Supports: comfortable speech below the ventilatory or lactate threshold.
31. This course's own constructions, each labelled in the body where it appears: why interval protocols are broken into bouts, and that the hard bouts are meant to go past the talk-test line; the reading that Milanović's 1.2 comes from direct comparisons rather than from subtracting the gains over controls; the made-up man whose 7 per cent gain is hidden by test error (only the ±3.96 per cent line comes from Montero and Lundby); adding up Couch to 5K's running minutes, 489 over nine weeks, and turning 17.8 injuries per 1,000 hours into roughly one beginner in seven; zone 2 as roughly the top of comfortable talking, joining Sitko's definition to Reed and Pipe's talk-test sentence; and the suggestion to test more than once at each end. The one-minute retelling in the opening is also this course's paraphrase: no news story or press release about the study was read.
32. G. A. Brooks, "The science and translation of lactate shuttle theory", Cell Metab 27(4), 2018, doi 10.1016/j.cmet.2018.03.008. Read level: abstract only. Supports: lactate as something muscles make continuously, not a waste product.
33. M. Pühringer and S. Ring-Dimitriou, "The influence of cardiorespiratory fitness level on the relationship between work rates at the aerobic threshold (AerT) and the point of maximal fat oxidation (Fatmax) in untrained adults", Front Sports Act Living 6, 2024, 1321896, doi 10.3389/fspor.2024.1321896. Read level: abstract only. Supports: 761 untrained adults aged 41 to 68, the two intensities and their weak agreement.