Mindful Minutes: What the Research Actually Shows About Journaling and Breathwork

A Stanford trial found five minutes of daily breathwork beat meditation on mood, while journaling's evidence is far more mixed. Here is what the randomized data supports, what it does not, and a protocol worth testing for 30 days.

Half of employees in the United States and Canada said they felt a lot of stress the previous day — the highest rate of any world region, according to Gallup's State of the Global Workplace: 2026 report, which put the global figure at 40 percent. Against that backdrop, "just breathe" sounds like a slogan. But breathwork and journaling have now been through enough randomized trials that we can talk about them in numbers rather than adjectives — including the numbers that are unflattering. The short version: five minutes of structured breathing has surprisingly good evidence behind it, journaling's evidence is genuinely mixed, and adherence is a bigger problem than efficacy for both.

Key statistics at a glance

50%
US and Canadian employees with a lot of stress the prior day
Gallup, 2026
5 min
Daily cyclic sighing that outperformed meditation on mood
Cell Reports Medicine, 2023
g = 0.81
Effect of slow-breathing biofeedback on stress and anxiety
Psychological Medicine, 2017
3.3%
Median 30-day retention across 93 mental-health apps
JMIR, 2019
  • 62% of US adults named societal division a significant source of stress, and 54% said they felt isolated from others — American Psychological Association, Stress in America 2025
  • g = −0.35 — the pooled effect of breathwork on self-reported stress across 12 randomized trials and 785 adults — Fincham et al., Scientific Reports (2023)
  • 47.1% vs 24.3% — the share of patients with asthma or rheumatoid arthritis showing clinically relevant improvement four months after three 20-minute writing sessions, versus controls — Smyth et al., JAMA (1999)
  • g = 0.04, not significant — the pooled effect of expressive writing on health across 30 randomized trials — Mogk et al., Psychosocial Medicine (2006)

Five minutes of breathing, measured against meditation

The most useful recent trial is also the most practical. Balban and colleagues at Stanford, publishing in Cell Reports Medicine in January 2023, randomized 108 adults to one of four five-minute daily practices for 28 days: cyclic sighing (a double inhale followed by a long, slow exhale), box breathing, cyclic hyperventilation with breath retention, or mindfulness meditation. One hundred participants completed the protocol.

All four groups improved. The breathing groups improved more on mood, and the exhale-weighted practice improved most.

Average daily improvement in positive affect after 28 days (5 min/day)
Cyclic hyperventilation1.97
Cyclic sighing1.89
Box breathing1.84
Mindfulness meditation1.22
Source: Balban et al., Cell Reports Medicine, 2023
Practice (5 min/day, 28 days)nAvg. daily change in positive affectAvg. daily change in state anxiety (STAI)
Cyclic sighing (double inhale, long exhale)27+1.89 ± 3.76 (p = 0.025)−3.85 ± 4.88
Box breathing (equal inhale/hold/exhale/hold)19+1.84 ± 3.24 (p = 0.026)−3.75 ± 2.83
Cyclic hyperventilation with retention32+1.97 ± 3.21 (p = 0.003)−1.81 ± 2.97
Mindfulness meditation22+1.22 ± 2.34 (p = 0.06)−3.95 ± 4.16

Two details matter more than the headline. First, cyclic sighing produced the largest drop in resting respiratory rate, and the size of that drop correlated with mood improvement (r = −0.24). The mechanism appears to be physiological, not just attentional. Second, cyclic sighing showed a significant interaction with cumulative days on protocol — the benefit compounded with practice rather than plateauing after the first session. Anxiety, notably, fell about equally in every group, meditation included. If your goal is less anxiety, the practice you will actually repeat is the right one.

The wider breathing evidence

One trial is one trial. The broader picture is consistent but more modest. A meta-analysis by Fincham and colleagues in Scientific Reports (2023) pooled 12 randomized controlled trials covering 785 adults and found breathwork lowered self-reported stress relative to controls with a small-to-medium effect (Hedges' g = −0.35, 95% CI −0.55 to −0.14, p = 0.0009). The authors rated most included studies at moderate risk of bias, which is the honest caveat.

The strongest signal comes from the most structured version of slow breathing. Heart rate variability biofeedback — typically paced breathing at roughly six breaths per minute, the frequency at which heart rate and breathing synchronize — showed a large effect on stress and anxiety across 24 studies and 484 participants in Goessl and colleagues' 2017 meta-analysis in Psychological Medicine: Hedges' g = 0.81 within-group and 0.83 versus controls. A separate systematic review of voluntary slow breathing, published in Neuroscience & Biobehavioral Reviews (2022), found large increases in RMSSD, a common heart rate variability measure, during slow-paced breathing compared with baseline.

The practical translation: pace matters more than technique branding. Roughly six breaths per minute — a five-second inhale and a five-second exhale — is the target most of this literature converges on, with extended exhales as the useful refinement.

What journaling can and cannot claim

Journaling's evidence is where careful reading pays off. The landmark trial is Smyth and colleagues' 1999 study in JAMA: 107 patients with asthma or rheumatoid arthritis wrote for 20 minutes on three consecutive days, either about the most stressful experience of their lives or about neutral topics. Four months later, asthma patients in the writing group improved from 63.9% to 76.3% of predicted FEV1 (a lung function measure) while controls did not change, and rheumatoid arthritis severity fell 28%. Overall, 47.1% of writers showed clinically relevant improvement versus 24.3% of controls.

That result is real, but it has not replicated cleanly at scale. A meta-analysis by Mogk and colleagues in Psychosocial Medicine (2006), restricted to 30 randomized controlled trials, found no significant pooled effect on somatic health (g = 0.07), psychological health (g = 0.01), or overall health (g = 0.04). Earlier, less restrictive meta-analyses reported effects around d = 0.19 to 0.47. Whatever expressive writing does, its average effect across all comers is small.

Structured, positive-focused writing has held up better in recent work. Smyth and colleagues' 2018 randomized trial in JMIR Mental Health assigned 70 medical patients with elevated anxiety to 15 minutes of online positive affect journaling three days a week for 12 weeks; the intervention group reported less distress, fewer depressive and anxiety symptoms after one month, and greater resilience relative to usual care. Gratitude writing shows a similar profile: a 2023 meta-analysis in the International Journal of Applied Positive Psychology pooled 25 randomized studies and 6,745 participants and found a small but reliable effect on psychological wellbeing (Hedges' g = 0.22, 95% CI 0.11–0.33), while a 2023 review of 64 randomized trials in einstein (São Paulo) concluded gratitude interventions improved mental health and reduced anxiety and depression symptoms.

On pen versus keyboard: a 2024 high-density EEG study in Frontiers in Psychology recorded 36 university students and found handwriting produced widespread theta-alpha connectivity across parietal and central brain regions that was absent during typing. A 2025 commentary in the same journal flagged real methodological limits — the study tested neither learning nor memory, and typing skill was not controlled. Treat handwriting as a reasonable preference, not a proven upgrade.

The real failure point is adherence

Efficacy is not the bottleneck. Use is. Meditation use among US adults roughly doubled from 7.5% in 2002 to 17.3% in 2022, per the National Health Interview Survey summarized by NCCIH — real growth, but still fewer than one in five adults. And the apps most people reach for do not hold anyone. Baumel and colleagues, analyzing real-world usage of 93 mental health apps in the Journal of Medical Internet Research (2019), found a median daily active user rate of 4.0%, median 15-day retention of 3.9%, and median 30-day retention of 3.3%. Mindfulness and meditation apps did comparatively well at 4.7%. Breathing-exercise apps had a median 30-day retention of 0.0%.

That is the number to design around. The Stanford protocol works partly because it is five minutes, unsupervised, and requires nothing. Anchoring it to an existing daily event — the end of a commute, the moment before the first meeting, lights-out — is doing more work than the technique choice. The same logic applies to sleep routines and to workplace burnout prevention, where consistency consistently beats intensity.

A 30-day protocol worth testing

PracticeDose used in researchBest-supported outcome
Cyclic sighing5 min daily, 28 daysPositive affect, lower respiratory rate
Slow-paced breathing (~6 breaths/min)10–20 min sessions, multi-weekStress and anxiety, heart rate variability
Expressive writing20 min, 3 consecutive daysSituational; small average effect
Positive affect journaling15 min, 3 days/week, 12 weeksDistress, anxiety, resilience
Gratitude writingWeekly entries, several weeksWellbeing (small effect, g = 0.22)

A defensible starting stack, drawn directly from the doses above: five minutes of cyclic sighing every day at a fixed time, plus 15 minutes of writing three days a week. Reserve the 20-minute expressive writing format for a specific unresolved stressor rather than making it the daily default — that is the context in which it was tested.

What to check at day 30

Track two things, not ten: whether you completed the practice (a simple yes/no per day) and a single one-to-ten stress rating taken at the same hour. The Stanford data suggests the mood benefit grows with cumulative days, so a four-week run is roughly the minimum honest test. Chronic stress also does not stay psychological — the American Heart Association's Mind-Heart-Body Connection scientific statement in Circulation (2021) concluded that psychological health is associated with cardiovascular health and disease risk, which is one reason stress habits belong in the same conversation as the rest of your long-term planning. If that planning includes protecting the people who depend on you, you can get a coverage quote in about the time one of these sessions takes.

Sources