How to read a health study in 5 minutes

Author: Raymond Hackland · Last reviewed: 27 October 2026 · Reading time: about 8 minutes

“Studies show…” You’ll hear it about breathing, about sound, about Qigong.

I once saw a flyer that said Qigong cures high blood pressure. That’s a red flag. But if you don’t know, you don’t know. I was curious, because I’ve felt the benefits myself. So I looked into the actual research. There were several studies, some of them promising, but most were of poor quality. And there’s a big difference between something helping and something curing. If Qigong could cure high blood pressure, I think we’d all be hearing a lot more about it. So be a little sceptical, and do some thinking for yourself. That’s what this article is for.

Here’s a simple routine for looking at any health study – five questions, one a minute. It opens my series “Reading the Evidence”.

The questions aren’t mine. They come from the tools researchers themselves use to judge studies, such as the Cochrane Handbook and the GRADE system [1, 2]. Five minutes won’t make you a reviewer. But they will catch the most common problems. And the next time someone tells you a breathing, sound or Qigong practice “has been shown” to do something, you can check for yourself how much that claim really rests on.

What science says

You don’t need a statistics degree to read a study. Most of what matters is in the abstract (the summary at the top) and the methods section. Researchers who write systematic reviews ask much the same questions, only in more detail [1, 2]. Here’s a five-minute version, with examples from studies the hub has already written about.

Minute 1: Who was studied, and how many? Look for the number of participants, usually written as “n”, and who they were. The 2002 study behind most humming claims tested ten healthy volunteers [3]. The often-quoted 528 Hz study had nine healthy adults, eight of them in the heart data [4]. The 432 Hz pilot study had 33 healthy volunteers [7]. That doesn’t make these studies wrong. But a handful of healthy people in one lab can’t tell you what happens in everyone, or in people who are ill. Example labels: Early evidence (humming) · Speculative (528 Hz and 432 Hz claims)

Minute 2: Compared with what? A result only means something next to a comparison. Good trials assign people to groups at random. That way, on average, the groups start out alike and differ mainly in the thing being tested [1]. In the 528 Hz study, no randomisation of the order was reported. Each music version was compared with its own baseline, not directly with the other [4]. Then there’s the 2020 review of Liu Zi Jue trials in chronic lung disease (COPD). There, the control groups received usual care. The authors themselves name methodological bias and possible placebo effects as reasons why more research is needed [5]. When people know what they’re getting, expectation can explain part of the difference. Example label: Early evidence (Liu Zi Jue)

Minute 3: What was actually measured? Many studies measure a body signal, not a health outcome. The humming study measured nitric oxide in the nose. It was 15 times higher during humming than during quiet exhalation [3]. That’s a striking laboratory finding. But it isn’t a measurement of fewer colds or healthier sinuses. Slow-breathing research mostly measures heart rate variability (HRV), a heart-rhythm signal. A 2022 meta-analysis of 223 studies found that the vagally mediated part of HRV rose during slow breathing [6]. So ask yourself: is the thing measured the thing the headline talks about? Example labels: Early evidence (humming) · Well studied (slow breathing and HRV)

Minute 4: How big is the effect, and how sure are they? A result can be “statistically significant” and still small. And a result can be not significant without showing that there’s no effect. In the 432 Hz pilot study, heart rate was on average about 4.8 beats per minute lower with 432 Hz than with 440 Hz. The p-value was 0.05, right at the usual significance threshold [7]. Look for an effect size and the range around it, the confidence interval. In a Cochrane review, tai chi reduced the number of older people who fell by about 20%. That’s a risk ratio of 0.80, with a 95% confidence interval of 0.70 to 0.91, rated high certainty [8]. A narrow range that stays clearly on one side of “no effect” is more convincing than a single p-value. Example labels: Speculative (432 Hz claims) · Well studied (tai chi and falls)

Minute 5: What else agrees, and who says so? One study is a starting point. Check whether others, ideally independent groups, found the same, and whether there’s a systematic review. GRADE is the system many review authors use. It rates the whole body of evidence for each outcome, not single studies, from high to very low certainty. It lowers its rating for risk of bias, inconsistent results, indirectness, imprecision and publication bias [2].

The humming studies come almost entirely from one research network, with no large independent replication, as the hub’s humming article explains. The tai chi and falls finding is different. It comes from a Cochrane review of 108 trials, which pooled 8 tai chi trials with 2,677 participants for this result [8]. The same review shows why GRADE rates each outcome separately. For the overall rate of falls, its tai chi evidence is only low certainty [8]. While you’re at it, glance at who funded the study, and whether the authors declare conflicts of interest.

Your five questions on one card:

• Who, and how many?
• Compared with what?
• What was measured?
• How big, and how sure?
• What else agrees?

The question I ask most is the first one: who was studied, and how many? You’d think bigger is always better, but it isn’t that simple. A study can have lots of people and still tell you very little. You can have a study with 10,000 healthy 25-year-olds and 10,000 80-year-olds on medication, and the average result might not tell you much about either group. So I always want to know who was actually studied, and how many of them.

Back to that flyer. Here’s the review I looked at when I saw it: a 2015 systematic review of Qigong for high blood pressure [9]. Let’s run the five questions.

• Who, and how many? 20 randomised trials with 2,349 people who had high blood pressure [9].
• Compared with what? No treatment, exercise, or blood pressure medication. Some trials compared Qigong plus medication with medication alone [9].
• What was measured? Blood pressure readings. Only one of the 20 trials looked at hard outcomes such as deaths [9].
• How big, and how sure? Compared with no treatment, blood pressure fell by about 17 mm Hg (systolic) and 10 mm Hg (diastolic). Qigong plus medication lowered it more than medication alone. Compared with exercise, Qigong did less for systolic pressure. But the authors rate the risk of bias in the trials as generally high [9].
• What else agrees? The authors themselves say better designed trials with long-term follow-up are needed to confirm the results [9].

A fair summary: promising, but not proof, and certainly not a cure. Helping and curing are two different things.

To see how all this turns into the labels on every article, read our evidence labels. Next in the series: Why one study isn’t proof (coming soon), on sample size, replication and p-values in plain words. Already in the series: Study types explained: from anecdote to meta-analysis and Spotting a misleading health headline. A glossary of research terms is also planned.

Where popular claims go beyond the evidence

Label: Speculative

• “A study shows…” One study rarely shows anything on its own. Ask which study, how many people, and compared with what.
• “Published in a journal, so it’s true.” Peer review checks a paper. It doesn’t guarantee the result. The 528 Hz study was published, and its own design limits what it can show [4].
• “Significant means important.” Statistical significance doesn’t measure how large or important an effect is.
• “It worked in the lab, so it works for you.” A result from ten healthy volunteers, measuring one gas [3], says nothing direct about your health.

On social media you’ll often read that studies show 528 Hz repairs your DNA. I know of no credible study that shows it repairs DNA, with a proper control group and enough people. But no study doesn’t mean the claim is false, and that’s something we should also keep in mind. It may just mean nobody has funded one, and good studies cost a lot. So a lack of evidence isn’t proof that something doesn’t work. It means we don’t know yet. That’s where I sit. Science is one side of the coin, with what’s been studied and confirmed so far. Experience is the other side. I’m the bit in the middle, trying to bridge the gap between the two sides.

How this relates to the Six Healing Sounds, Qigong and neigong

Labels: Early evidence (research on Liu Zi Jue) · Tradition only (qi, neigong concepts) · my own approach is marked as mine, and its link to the science of vibration is Speculative

Let’s try the five questions on a typical claim: “Studies show the Six Healing Sounds improve lung function.”

• Who was studied? Mostly people with stable COPD, not healthy practitioners [5].
• Compared with what? Usually usual care, with no placebo-like comparison [5].
• What was measured? Things like breathlessness, walking distance, lung function and quality of life [5].
• How sure? The review authors call for more research because of bias and possible placebo effects [5].
• What else agrees? Several reviews report similar results. But they probably share many of the same trials, as the hub’s article on what systematic reviews say about Qigong and tai chi explains.

A fair summary: encouraging early results in one patient group, studied alongside medical care.

The five questions also show where research stops. Qi, the organ pairings of the six sounds and the inner work of neigong are traditional concepts. Research neither confirms nor refutes them, and they’re probably not testable as stated. My articles on the history of Liu Zi Jue, what Qigong is and what neigong is give that background as tradition, not as findings.

How my practice relates to this. My Six Healing Sounds practice is my own, and it’s breath-led. You breathe in quietly through the nose and breathe the six sounds out through the mouth – “voiced through the breath”, not sung – working over time towards a silent exhalation. The breathing stays slow and controlled throughout, with no breath holds and no fast breathing.

Speculative, my own idea: This part is my own speculation. I link the sounds to the science of sound and vibration. Every object has its own natural frequencies, so in theory my lungs, my liver and my other organs each have their own frequency too. If I could match that frequency with my breath, I might be able to connect with that organ. But I don’t know what frequency my organs have, and even if I did, I don’t know that it would work. Maybe that’s where neigong comes in: the intention and the visualisation. I still don’t know if it works, but it feels like it does. What can be felt is the vibration of your own voice, as What is resonance? explains. And I know of no measurements of a particular sound vibrating a particular organ (see the hub’s article on vowels and formants).

I also use a visualisation in the spirit of neigong. You imagine yourself as the resonance, travelling with each sound to its organ, and with xi through the whole body (the triple burner). In this visualisation, “resonance” is an image, not a physical claim. When I breathe the sounds instead of singing or humming them, it all feels lighter. It’s less about the voice and more a feeling of connectedness. That’s my own sense of it, not a measurement.

This rests on my own experience and that of participants in my Immersions, not on a scientific finding. I care about both: what the research shows and what people actually feel. Measured against the five questions, experience like this counts as anecdote – worth hearing, but not a study. None of this is a health claim. If you have a health condition, please read who should be careful with breathing practices first.

Resonance Heritage Qi is a personal, music-assisted practice inspired by Liu Zi Jue. It’s connected to it, not a replacement for the traditional form or for study with a teacher of that tradition. No study has tested it, and I make no claims about its effects.

So next time you read “studies show”, give it five minutes. Who, compared with what, what was measured, how big, and what else agrees.

So read these articles, but be sceptical, and ask questions.

Sources

1. Higgins JP, Savović J, Page MJ, Elbers RG, Sterne JAC, 2019 (last updated October 2019). Chapter 8: Assessing risk of bias in a randomized trial. In: Higgins JP, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, et al. (eds). Cochrane Handbook for Systematic Reviews of Interventions, version 6.5. Cochrane, 2024. https://training.cochrane.org/handbook/current/chapter-08
2. Balshem H, Helfand M, Schünemann HJ, Oxman AD, Kunz R, Brozek J, Vist GE, Falck-Ytter Y, Meerpohl J, Norris S, Guyatt GH, 2011. GRADE guidelines: 3. Rating the quality of evidence. Journal of Clinical Epidemiology 64(4):401–406. https://doi.org/10.1016/j.jclinepi.2010.07.015
3. Weitzberg E, Lundberg JO, 2002. Humming greatly increases nasal nitric oxide. American Journal of Respiratory and Critical Care Medicine 166(2):144–145. https://doi.org/10.1164/rccm.200202-138BC
4. Akimoto K, Hu A, Yamaguchi T, Kobayashi H, 2018. Effect of 528 Hz music on the endocrine system and autonomic nervous system. Health 10(9):1159–1170. https://doi.org/10.4236/health.2018.109088
5. Xiao L, Duan H, Li P, Wu W, Shan C, Liu X, 2020. A systematic review and meta-analysis of Liuzijue in stable patients with chronic obstructive pulmonary disease. BMC Complementary Medicine and Therapies 20:308. https://doi.org/10.1186/s12906-020-03104-1
6. Laborde S, Allen MS, Borges U, Dosseville F, Hosang TJ, Iskra M, Mosley E, Salvotti C, Spolverato L, Zammit N, Javelle F, 2022. Effects of voluntary slow breathing on heart rate and heart rate variability: a systematic review and a meta-analysis. Neuroscience & Biobehavioral Reviews 138:104711. https://doi.org/10.1016/j.neubiorev.2022.104711
7. Calamassi D, Pomponi GP, 2019. Music tuned to 440 Hz versus 432 Hz and the health effects: a double-blind cross-over pilot study. Explore (NY) 15(4):283–290. https://doi.org/10.1016/j.explore.2019.04.001 (corrigendum: Explore 2020;16(1):8, https://doi.org/10.1016/j.explore.2020.01.001)
8. Sherrington C, Fairhall NJ, Wallbank GK, Tiedemann A, Michaleff ZA, Howard K, Clemson L, Hopewell S, Lamb SE, 2019. Exercise for preventing falls in older people living in the community. Cochrane Database of Systematic Reviews 1:CD012424. https://doi.org/10.1002/14651858.CD012424.pub2
9. Xiong X, Wang P, Li X, Zhang Y, 2015. Qigong for hypertension: a systematic review. Medicine (Baltimore) 94(1):e352. https://doi.org/10.1097/MD.0000000000000352

Source quality note: [1] is the current Cochrane methods handbook and [2] the GRADE working group’s guidance on rating evidence. Both are standard references in evidence-based medicine. [3]–[8] are the studies the hub’s earlier articles discuss, used here only as worked examples. Their numbers match those articles. We rechecked [4], [5] and [8] in the full texts and [3], [6] and [7] in their published abstracts. We could not open the corrigendum to [7]. [8] is a Cochrane review, [5] and [6] are systematic reviews, and [3], [4] and [7] are small single studies. [9] is a systematic review added as my own worked example, checked in the full text.