Chapter 8 Practice and a Pre-registered Trial

Note

Please reread the health notice at the start of the preface. The checklist in this chapter is a tool for recording and thinking, not an exercise prescription, a nutrition plan or medical advice. Before starting a new exercise programme, assess your situation with the American College of Sports Medicine’s preparticipation health screening; anyone with heart disease, high blood pressure, diabetes, kidney disease or pregnancy, or who has had chest pain, dizziness, fainting or palpitations during exercise, should consult a doctor first. Overnight fasts in the checklist never exceed 16 hours, and the groups listed as unsuitable in Books 5 and 17 should not lengthen their fasting windows on their own. This chapter gives no doses of any drug or supplement and sets no weight target.

Part 1 A four-step physiological resilience and recovery checklist

This checklist turns several models from earlier chapters into four questions that can be recorded each week and compared with what happens later. Its purpose is not to “optimise” the body but to let a person see clearly what they are doing and how the body responds, and to stop when warning signs appear.

Weekly Zone 2 minutes and HRR60. Record the minutes of aerobic activity each week below the heart rate of the first ventilatory threshold (if you have not had a supervised exercise test, use a “talking” intensity at which you can speak in full sentences). Once a week, under the same conditions, do the same standard submaximal effort (for example 6 minutes of walking at a fixed speed and grade), then stand still and record the heart rate at the moment you stop and 60 seconds later; the difference is HRR60, the one-minute heart-rate recovery. If HRR60 is 12 beats per minute or less, do not interpret it yourself; consult a doctor (Part 2 explains where this figure comes from). (Chapters 3, 5 and 6)
Eating and fasting window audit. Record the time of the last meal each day and of the first meal the next day, and compute the overnight fast; this book’s upper limit is 16 hours, and days above it are flagged. The purpose of this step is to spot irregularity and excess, not to pursue longer fasts: Chapter 2 showed that the effect of fasting windows on tissue damage has never been measured in humans. (Chapter 2)
Stress pulses against sleep recovery. Record each week’s “high-intensity pulses” (interval sessions, long hill climbs, races and so on) and the number of nights with at least 7 hours’ sleep. The book uses a conservative rule of thumb: at least two nights of adequate sleep for every high-intensity pulse, and for beginners no high-intensity pulses on consecutive days. The rule comes from the qualitative conclusions of the model in Chapter 3; it is not a threshold established by trials. (Chapters 3 and 6)
Checking the evidence grade of outside interventions. For every “anti-ageing” drug, supplement or method you use or plan to use, write down the specific effect claimed, then find the grade of the best evidence for that effect: A for human randomised trials with that effect as the outcome, B for human randomised trials with markers only as outcomes, C for human observational studies, D for animal or cell studies, E for mechanistic reasoning or personal experience. Prescription drugs are never to be self-prescribed, and the book gives no doses. (Chapters 2, 3, 5 and 7)

Calculation

The checklist applied to a fictional four-week log kept by a person called L, who starts with about 95 minutes of low-intensity walking a week. All numbers are invented.

Calculation

# Four-step checklist on a FICTIONAL four-week log (person L). Rules: HRR60 <= 12 -> ask a doctor;
# overnight fast cap 16 h; >= 2 nights of >= 7 h sleep per hard session, none on consecutive days.
WEEKS = [  # zone-2 min, HRR60 bpm, hard-session days (0 = Mon), nights >= 7 h sleep
    (95, 17, (2,), 5), (120, 18, (1, 2), 4), (140, 19, (1, 4), 5), (150, 21, (1, 3, 5), 4)]
FASTS = [12.5, 13, 14, 16.5, 13, 12, 14, 15, 13.5, 12, 14, 17, 13, 12,
         14, 13, 12.5, 14, 15.5, 12, 13, 13, 14, 12.5, 16, 13, 14, 12]   # hours, 28 nights

hrr = [w[1] for w in WEEKS]
print("step 1  zone-2 min/week: " + ", ".join(str(w[0]) for w in WEEKS) +
      "; HRR60: " + ", ".join(map(str, hrr)) + (" (<= 12: ask a doctor)" if min(hrr) <= 12 else ""))
mx, my = 1.5, sum(hrr) / 4
slope = sum((x - mx) * (y - my) for x, y in enumerate(hrr)) / 5.0
print(f"        HRR60 trend {slope:+.1f} bpm/week over 4 weeks (one person, no control: not proof)")
over = [i + 1 for i, h in enumerate(FASTS) if h > 16]
print(f"step 2  overnight fast: mean {sum(FASTS) / 28:.1f} h, longest {max(FASTS)} h;"
      f" nights above the 16 h cap: {over}")
for i, (_, _, hard, nights) in enumerate(WEEKS, 1):
    flags = ([f"only {nights / len(hard):.1f} good nights per hard session"]
             if nights / len(hard) < 2 else []) + \
            (["hard sessions on consecutive days"] if any(b - a == 1 for a, b in
                                                          zip(hard, hard[1:])) else [])
    print(f"step 3  week {i}: {len(hard)} hard, {nights} nights >= 7 h -> "
          + ("; ".join(flags) or "ok"))
GRADE = {"RCT of the claimed outcome": "A", "RCT of a marker only": "B",
         "human observational": "C", "animal or cell": "D", "mechanism or anecdote": "E"}
ITEMS = [  # (intervention, claim checked, best evidence found FOR THAT CLAIM, action)
    ("brisk walking / aerobic training", "raises fitness", "RCT of the claimed outcome",
     "reasonable to continue"),
    ("prescription drug sold as anti-ageing", "extends human life", "animal or cell",
     "do not self-prescribe; ask a doctor"),
    ("supplement said to raise NAD+", "slows human ageing", "animal or cell", "no reason to act"),
    ("regular sauna bathing", "lowers mortality", "human observational", "no reason to act"),
    ("'detox' cold plunge", "switches on autophagy", "mechanism or anecdote", "no reason to act")]
for name, claim, best, act in ITEMS:
    print(f"step 4  {name}, claim '{claim}'")
    print(f"        best evidence {GRADE[best]} ({best}) -> {act}")
# step 1  zone-2 min/week: 95, 120, 140, 150; HRR60: 17, 18, 19, 21
#         HRR60 trend +1.3 bpm/week over 4 weeks (one person, no control: not proof)
# step 2  overnight fast: mean 13.6 h, longest 17 h; nights above the 16 h cap: [4, 12]
# step 3  week 1: 1 hard, 5 nights >= 7 h -> ok
# step 3  week 2: 2 hard, 4 nights >= 7 h -> hard sessions on consecutive days
# step 3  week 3: 2 hard, 5 nights >= 7 h -> ok
# step 3  week 4: 3 hard, 4 nights >= 7 h -> only 1.3 good nights per hard session
# step 4  brisk walking / aerobic training, claim 'raises fitness'
#         best evidence A (RCT of the claimed outcome) -> reasonable to continue
# step 4  prescription drug sold as anti-ageing, claim 'extends human life'
#         best evidence D (animal or cell) -> do not self-prescribe; ask a doctor
# step 4  supplement said to raise NAD+, claim 'slows human ageing'
#         best evidence D (animal or cell) -> no reason to act
# step 4  regular sauna bathing, claim 'lowers mortality'
#         best evidence C (human observational) -> no reason to act
# step 4  'detox' cold plunge, claim 'switches on autophagy'
#         best evidence E (mechanism or anecdote) -> no reason to act

Step 1 shows L’s Zone 2 minutes rising from 95 to 150 a week and HRR60 from 17 to 21 beats per minute (about +1.3 a week); but one person, four data points and no control, with sleep, caffeine, temperature and measurement error all at work, cannot show that the training “works”. Step 2 flags nights 4 and 12, above the 16-hour limit (16.5 and 17 hours), whose causes L should look into. Step 3 flags week 2, with two high-intensity pulses on consecutive days, and week 4, with three pulses matched by only four nights of adequate sleep (1.3 per pulse).

Step 4 is a reminder that “supported by research” can mean very different grades. That aerobic training raises fitness rests on randomised trials with fitness as the outcome (A); for a prescription drug promoted as “anti-ageing” and a supplement said to raise NAD⁺, the best evidence for “extends life” or “slows ageing” comes from animals (D); sauna bathing has only an observational link with lower mortality (C); and “a cold plunge that switches on autophagy” rests on mechanism and anecdote alone (E).

Note

A fictional log is not evidence. The numbers above only show how the checklist computes and flags. The checklist’s value lies in writing judgements down so that they can be compared with what happens later, not in calculating a “correct” amount of training. Chest pain, unusual breathlessness, dizziness or palpitations in any week call for stopping and seeing a doctor, not for adjusting the checklist.

Part 2 A pre-registered randomised controlled trial

Question

In sedentary adults, does 12 weeks of supervised incline treadmill walking, three 45-minute sessions a week at an intensity kept below the first ventilatory threshold, speed up heart-rate recovery one minute after submaximal exercise (HRR60) compared with standard advice on daily steps?

Evidence

Why HRR60. The fall in heart rate in the first minute after exercise stops mainly reflects the reactivation of the parasympathetic nervous system. Cole and colleagues (1999) followed 2,428 patients who had exercise testing for clinical reasons for 6 years, during which 213 died; defining a fall of 12 beats per minute or less in the first minute after exercise as abnormal, 6-year mortality was 19 % with abnormal recovery and 5 % with normal recovery, and abnormal heart-rate recovery remained an independent predictor of death in multivariable analysis (Cole et al. 1999). The threshold comes from symptom-limited maximal tests and the recovery protocol that followed them, and cannot be carried over directly to the self-administered submaximal test of Part 1; the checklist uses it only as a conservative prompt to consult a doctor, not as a diagnostic criterion. This is an observational study: HRR60 is a prognostic marker, and raising it is not the same as lowering mortality. On the other side, a systematic review of the effect of exercise training on cardiac parasympathetic activity in sedentary people found too few data to meta-analyse the effect of training on HRR (Casanova-Lizón et al. 2022). The trial therefore cannot borrow a published effect size, and its smallest effect of interest is set in advance.

Safety: preparticipation screening and an assessment of the “15 % grade”

Evidence

Riebe and colleagues (2015) updated the American College of Sports Medicine’s recommendations for preparticipation health screening (Riebe et al. 2015). The new process decides whether medical clearance is needed before exercise from three factors (whether the person currently exercises regularly; whether they have known cardiovascular, metabolic or renal disease; and whether they have signs or symptoms suggesting such disease) together with the planned exercise intensity. People who do not exercise regularly and have neither disease nor symptoms may begin at light to moderate intensity; those with known disease or any suggestive symptoms are advised to obtain medical clearance first.

The author’s draft envisaged treadmill walking at a 15 % grade. The program below uses the American College of Sports Medicine’s walking equation (American College of Sports Medicine 2021) to estimate energy expenditure at different speeds and grades and compares it with the first ventilatory threshold of three hypothetical sedentary adults (with the threshold assumed to lie at 55 % of maximal oxygen uptake; in the trial it is measured for each person).

Calculation

# ACSM walking equation: VO2 = 3.5 + 0.1 S + 1.8 S G (S in m/min, G as a fraction); 1 MET = 3.5.
# VT1 assumed at 55 % of VO2max for three illustrative sedentary adults.
MPH = 26.8224                                       # m/min per mile per hour

def vo2(mph, grade_pct):
    s = mph * MPH
    return 3.5 + 0.1 * s + 1.8 * s * grade_pct / 100

GRADES = (0, 5, 10, 12, 15)
print("  speed  | METs at grade:" + "".join(f"{g:6d} %" for g in GRADES))
for mph in (2.0, 2.5, 3.0, 3.5):
    print(f" {mph:3.1f} mph |" + " " * 14 + "".join(f"{vo2(mph, g) / 3.5:8.1f}" for g in GRADES))
print(" VO2max | VT1 (55 %) | 15 % grade at 3.0 mph | highest grade at VT1: 2.0 / 2.5 / 3.0 mph")
for vmax in (22.0, 28.0, 35.0):
    vt1 = 0.55 * vmax
    pct = vo2(3.0, 15) / vmax
    best = []
    for mph in (2.0, 2.5, 3.0):
        ok = [g / 2 for g in range(0, 41) if vo2(mph, g / 2) <= vt1]
        best.append(f"{max(ok):4.1f} %" if ok else "  none")
    print(f"  {vmax:4.1f}  |   {vt1:4.1f}     |  {pct:4.0%} of VO2max"
          f"         |  " + " / ".join(best))
#   speed  | METs at grade:     0 %     5 %    10 %    12 %    15 %
#  2.0 mph |                   2.5     3.9     5.3     5.8     6.7
#  2.5 mph |                   2.9     4.6     6.4     7.1     8.1
#  3.0 mph |                   3.3     5.4     7.4     8.3     9.5
#  3.5 mph |                   3.7     6.1     8.5     9.5    10.9
#  VO2max | VT1 (55 %) | 15 % grade at 3.0 mph | highest grade at VT1: 2.0 / 2.5 / 3.0 mph
#   22.0  |   12.1     |  151% of VO2max         |   3.0 % /  1.5 % /  0.0 %
#   28.0  |   15.4     |  119% of VO2max         |   6.5 % /  4.0 % /  2.5 %
#   35.0  |   19.2     |   95% of VO2max         |  10.5 % /  7.0 % /  5.0 %

Walking at 3.0 miles per hour (about 4.8 km/h) up a 15 % grade costs about 9.5 METs (33 mL/kg/min): 151 %, 119 % and 95 % of the maximum of people with maximal oxygen uptakes of 22, 28 and 35 mL/kg/min, all far above the first ventilatory threshold. To stay below it, the person with a maximum of 28 could use only about a 4 % grade at 2.5 miles per hour (about 4.0 km/h); the person with 35 up to about 10.5 % at 2.0 miles per hour (about 3.2 km/h).

Note

A correction of the author’s framework. A fixed 15 % grade is neither safe for most sedentary adults nor consistent with the design’s intention of staying “below the first ventilatory threshold”; steep walking also loads the calves and Achilles tendons considerably, and walking while holding the handrails makes the real load hard to estimate. The protocol therefore lets the investigators set grade and speed individually for each participant from the heart rate at the first ventilatory threshold measured at enrolment and from a talk test, starting at a low grade and raising it by at most 2 percentage points a week, with 15 % only as an upper limit; the handrails are never used to support body weight.

Design

Element	Specification
Participants	Sedentary adults aged 30–65 (under 150 minutes a week of moderate activity for 3 months, no structured training); not recruited from the author’s community, students, customers or card users, nor by targeting readers of the websites run by the author’s group
Screening	ACSM preparticipation screening (Riebe et al. 2015); physician-supervised, symptom-limited cardiopulmonary exercise test with 12-lead ECG, also measuring the first ventilatory threshold; anyone meeting criteria to stop the test is not enrolled and is referred
Exclusions	Known cardiovascular, cerebrovascular or peripheral vascular disease; uncontrolled hypertension; diabetes or other metabolic disease; kidney disease; pregnancy or planned pregnancy; beta-blockers or other drugs affecting heart rate; joint problems limiting walking; recent surgery or acute illness
Randomisation	1:1, stratified by sex and age band, random blocks; concealed allocation revealed by an independent data centre after baseline
Arm A	12 weeks, three 45-minute supervised incline walks a week (5-minute warm-up, 35 at target, 5 cool-down); heart rate at or below the first-threshold heart rate, able to talk in full sentences; grade and speed individualised and adjusted weekly
Arm B	One 20-minute standard education session (more daily steps per guidelines, written material); calls in weeks 4 and 8 without training guidance
Blinding	Participants and trainers unblinded; outcome assessors and statistician blinded; heart rate recorded automatically by chest strap
Primary outcome (single)	HRR60 at week 12: standard submaximal test with the same speed and grade stages at baseline and week 12 (ending at the stage that reached about 80 % of peak heart rate at baseline); standing still afterwards, HRR60 = heart rate at stopping minus heart rate 60 seconds later
Secondary	Peak oxygen uptake and oxygen uptake at the first threshold; resting heart rate; heart rate at the end of the submaximal stage; 7-day accelerometer steps; adherence; adverse events
Exploratory	Resting metabolic flexibility ΔRER: ventilated-hood indirect calorimetry after an overnight fast (within 16 hours); mean respiratory exchange ratio 60–120 minutes after a standardised mixed meal minus the fasting value (Galgani et al. 2008); affected by diet and energy balance, exploratory only
Analysis	Intention to treat; analysis of covariance (baseline HRR60, stratification factors); multiple imputation; two-sided $\alpha = 0.05$ ; sensitivity analyses adding heart rate at test end, and completers only
Safety	ACSM stopping criteria; serious adverse events reported within 24 hours to an independent data and safety monitoring board, with training suspended and referral

Note

Why the primary test uses “the same absolute workload”. After training, the same speed and grade produce a lower heart rate, and the heart rate at stopping itself affects HRR60. Using the same absolute workload measures “how fast one recovers after the same amount of work”, but it mixes “a lower heart rate during work” with “faster recovery”. The protocol therefore pre-specifies a sensitivity analysis with heart rate at the end of the test as a covariate. If the two analyses disagree, the report must present both.

Sample size

Calculation

# Sample size, 1:1 RCT, primary outcome HRR60 at week 12 (ANCOVA on baseline). Assumed: SESOI
# 4 bpm, SD 10 bpm, r = 0.6; alpha 0.05 two-sided; normal approximation + 1; 15 % loss.
import math
from statistics import NormalDist
z = NormalDist().inv_cdf

def per_arm(delta, sd, r, power=0.90, alpha=0.05):
    resid = sd * math.sqrt(1 - r * r)
    return math.ceil(2 * (z(1 - alpha / 2) + z(power)) ** 2 * resid ** 2 / delta ** 2) + 1

print(" effect (bpm) | r = 0.4 | r = 0.6 | r = 0.8   (analysable per arm, SD 10, 90 % power)")
for d in (3, 4, 5, 6):
    print(f"     {d:2d}       |  {per_arm(d, 10, 0.4):4d}   |  {per_arm(d, 10, 0.6):4d}   |"
          f"  {per_arm(d, 10, 0.8):4d}")
n = per_arm(4, 10, 0.6); n80 = per_arm(4, 10, 0.6, power=0.80)
recruit = math.ceil(n / 0.85)
print(f"chosen: 4 bpm, SD 10, r 0.6, 90 % power -> {n} analysable per arm (80 % power: {n80})")
print(f"allowing 15 % loss: randomise {recruit} per arm, {2 * recruit} in total")
se = 10 * math.sqrt(1 - 0.36) * math.sqrt(2 / n); P = NormalDist().cdf   # SE of the difference
print(f"SE of the difference {se:.2f} bpm; true effect 4: P(lower 95 % limit > 0) = "
      f"{P(4 / se - 1.96):.0%}, P(lower limit >= 4) = {P(-1.96):.1%}")
print(f"a 90 % chance that the lower limit reaches 4 needs a true effect of"
      f" {4 + (1.96 + z(0.90)) * se:.1f} bpm")
#  effect (bpm) | r = 0.4 | r = 0.6 | r = 0.8   (analysable per arm, SD 10, 90 % power)
#       3       |   198   |   151   |    86
#       4       |   112   |    86   |    49
#       5       |    72   |    55   |    32
#       6       |    51   |    39   |    23
# chosen: 4 bpm, SD 10, r 0.6, 90 % power -> 86 analysable per arm (80 % power: 64)
# allowing 15 % loss: randomise 102 per arm, 204 in total
# SE of the difference 1.22 bpm; true effect 4: P(lower 95 % limit > 0) = 91%, P(lower limit >= 4) = 2.5%
# a 90 % chance that the lower limit reaches 4 needs a true effect of 8.0 bpm

With a smallest effect of interest of 4 beats per minute, a standard deviation of HRR60 of 10 beats per minute and a correlation of 0.6 between baseline and week 12 (a residual standard deviation of 8 beats per minute after analysis of covariance), two-sided $\alpha = 0.05$ and 90 % power, each arm needs 86 analysable participants; allowing for about 15 % loss to follow-up, 102 are randomised per arm, 204 in all. None of these assumptions can be taken directly from a published estimate: with a correlation of only 0.4, each arm would need 112; if the true effect were only 3 beats per minute, 151. The protocol publishes this sensitivity table at pre-registration.

It should be clear what the 90 % power refers to. If the true effect is exactly 4 beats per minute, the chance that the lower confidence limit lies above 0 (the verdict “speeds HRR60”) is about 91 %; the chance that it reaches 4, allowing the statement “by at least the smallest effect of interest”, is only 2.5 %, and a 90 % chance of the stronger statement would need a true effect of about 8 beats per minute. At this sample size, then, even a real effect of the expected size will most probably be reported as “speeds HRR60; size not shown to reach 4 bpm”.

Decision rule

Calculation

# Decision rule fixed before enrolment: diff = A - B in adjusted HRR60 (bpm), 95 % CI; SESOI 4.
SESOI = 4.0

def verdict(diff, lo, hi, adherence_ok=True, intensity_ok=True, blind_ok=True):
    if not blind_ok:
        return f"diff {diff:+.1f}: NOT INTERPRETABLE - outcome assessors unblinded"
    if not adherence_ok:
        return f"diff {diff:+.1f}: NOT INTERPRETABLE - < 70 % of arm A attended >= 75 % of sessions"
    if not intensity_ok:
        return f"diff {diff:+.1f}: NOT INTERPRETABLE - training was not at the intended intensity"
    if lo >= SESOI:  v = "speeds HRR60 by at least the smallest effect of interest"
    elif lo > 0:     v = "speeds HRR60; size not shown to reach 4 bpm"
    elif hi < 0:     v = "SLOWS HRR60"
    elif lo >= -SESOI and hi <= SESOI: v = "no difference worth acting on"
    else:            v = "unclear: the interval is too wide to decide"
    return f"diff {diff:+.1f} bpm ({lo:+.1f} to {hi:+.1f}): {v}"

for case in ((7.0, 4.6, 9.4), (4.5, 2.1, 6.9), (0.8, -1.6, 3.2), (-3.0, -5.4, -0.6),
             (1.5, -4.5, 7.5), (6.0, 3.6, 8.4, False), (6.0, 3.6, 8.4, True, False),
             (6.0, 3.6, 8.4, True, True, False)):
    print(verdict(*case))
print("secondary and exploratory outcomes (VO2peak estimate, resting heart rate, metabolic")
print("flexibility as delta-RER, daily steps) are reported with intervals and never decide")
# diff +7.0 bpm (+4.6 to +9.4): speeds HRR60 by at least the smallest effect of interest
# diff +4.5 bpm (+2.1 to +6.9): speeds HRR60; size not shown to reach 4 bpm
# diff +0.8 bpm (-1.6 to +3.2): no difference worth acting on
# diff -3.0 bpm (-5.4 to -0.6): SLOWS HRR60
# diff +1.5 bpm (-4.5 to +7.5): unclear: the interval is too wide to decide
# diff +6.0: NOT INTERPRETABLE - < 70 % of arm A attended >= 75 % of sessions
# diff +6.0: NOT INTERPRETABLE - training was not at the intended intensity
# diff +6.0: NOT INTERPRETABLE - outcome assessors unblinded
# secondary and exploratory outcomes (VO2peak estimate, resting heart rate, metabolic
# flexibility as delta-RER, daily steps) are reported with intervals and never decide

The example results are invented. Only if the lower limit of the 95 % confidence interval is at least 4 beats per minute will the report state that “incline walking sped up HRR60 by at least the smallest effect of interest”. If fewer than 70 % of arm A completed at least 75 % of their sessions as specified, or more than 25 % of training time was above the heart-rate ceiling of the first ventilatory threshold, or the blinding of outcome assessors was broken, the result is “not interpretable” whatever it shows. Serious adverse events are reported prominently whatever the effect.

What this trial cannot show

First, HRR60 is a prognostic marker; speeding it up is not the same as lowering mortality or extending life, and the trial cannot answer the causal question of Chapter 5, how much raising fitness lowers mortality. Second, the trial compares “supervised training” with “one education session”, so the arms differ in their contact time with staff, and a positive result cannot be attributed to incline walking alone. Third, participants are sedentary adults who passed screening and were willing to train for 12 weeks; the results cannot be extended to people with cardiovascular disease, older people or athletes. Fourth, 12 weeks is too short to say anything about long-term adherence or long-term effects. Fifth, the exploratory metabolic-flexibility outcome depends heavily on diet and energy balance; even if it differs between arms, it cannot show what happened to “mitochondria” or “autophagy”, which would need flux measurements in tissue (Chapter 2).

Ethics and conflicts of interest

The trial must be approved by an independent ethics committee, participants give informed consent, and the protocol and analysis plan are registered before the first participant is enrolled (Nosek et al. 2018). Participants may withdraw at any time without giving a reason; the control arm is offered the same supervised training after the trial. The protocol includes no drug, supplement or prolonged-fasting intervention.

Disclosure

The author uses the works of 5D Library, including this book, as the methodological basis of his community courses, the 5D Ascension Cards (Dimensional Shift Card Decks) and the Yafu AI knowledge base, and therefore has a commercial interest in the subject of this book. In addition, the author and the group that includes 5D Library operate ketohackers.com and nadplusnadplus.com, popular-science websites about ketogenic eating and NAD⁺ respectively, and therefore have a direct commercial interest in the subject of this book; the book names no brand or product, recommends no supplement and gives no doses. The book promotes no course, card deck or service; the Chapter 8 trial is designed to be run and analysed by researchers independent of the author’s business and does not recruit from the author’s community, students or customers; all the code and the study protocol are published openly for anyone to check.

To limit the influence of these interests, the trial materials, analysis code and decision rule are published before enrolment, and the results, data and code will be published whatever the outcome.

References

American College of Sports Medicine. 2021. ACSM’s Guidelines for Exercise Testing and Prescription. 11th ed. Edited by Gary Liguori. Wolters Kluwer.

Casanova-Lizón, A., A. Manresa-Rocamora, A. A. Flatt, J. M. Sarabia, and M. Moya-Ramón. 2022. “Does Exercise Training Improve Cardiac-Parasympathetic Nervous System Activity in Sedentary People? A Systematic Review with Meta-Analysis.” International Journal of Environmental Research and Public Health 19 (21): 13899. https://doi.org/10.3390/ijerph192113899.

Cole, Christopher R., Eugene H. Blackstone, Fredric J. Pashkow, Claire E. Snader, and Michael S. Lauer. 1999. “Heart-Rate Recovery Immediately After Exercise as a Predictor of Mortality.” New England Journal of Medicine 341 (18): 1351–57. https://doi.org/10.1056/NEJM199910283411804.

Galgani, Jose E., Cedric Moro, and Eric Ravussin. 2008. “Metabolic Flexibility and Insulin Resistance.” American Journal of Physiology: Endocrinology and Metabolism 295 (5): E1009–17. https://doi.org/10.1152/ajpendo.90558.2008.

Nosek, Brian A., Charles R. Ebersole, Alexander C. DeHaven, and David T. Mellor. 2018. “The Preregistration Revolution.” Proceedings of the National Academy of Sciences 115 (11): 2600–2606. https://doi.org/10.1073/pnas.1708274114.

Riebe, Deborah, Barry A. Franklin, Paul D. Thompson, et al. 2015. “Updating ACSM’s Recommendations for Exercise Preparticipation Health Screening.” Medicine & Science in Sports & Exercise 47 (11): 2473–79. https://doi.org/10.1249/MSS.0000000000000664.

Fan, Y. (2026). Ageing Kinetics, the mTOR-AMPK Autophagy Switch and Hormesis: From the Gompertz Mortality Law and Mitohormesis to Cardiorespiratory Fitness Epidemiology, Read Beside Classical Chinese Longevity Philosophy [衰老动力学、mTOR-AMPK自噬开关与毒物兴奋效应的生物数学模型] (Version 1.0.0). 5D Library. https://doi.org/10.5281/zenodo.23770092 · Licensed under Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0).