§ Claim under review · Health
"A study found that eating 2 kiwifruits 1 hour before bedtime for 4 weeks led to: +1 hour of total sleep time; 47% reduction in wake time after sleep onset; Better overall sleep quality. Kiwis are naturally rich in serotonin and antioxidants — both linked to improved sleep patterns. Source: Doherty R., Madigan S., Nevill A., Warrington G., Ellis J.G. (2023)."
Verdict
Source exists but framing is misleading
Confidence
MediumSummary
The study cited in this post is real. Doherty and colleagues published it in the journal Nutrients in 2023, and it did find that eating two kiwifruit an hour before bed for four weeks was followed by more total sleep time, less time awake during the night, and better self-reported sleep quality. Total sleep time rose from about 7.6 hours to about 8.55 hours, so the "one hour" figure is roughly right, though the authors report the significant increase as about 50 minutes. What the post leaves out matters a great deal. The study included only 15 elite athletes, had no control group and no placebo, everyone knew they were eating kiwifruit, and all sleep data came from participants filling in their own sleep diaries rather than any objective measurement. That design cannot show that the kiwifruit caused the improvement, and the participants were already sleeping over seven and a half hours a night, so this is not evidence about people with insomnia. Eating kiwifruit before bed is harmless and cheap to try, but the evidence behind this claim is a small preliminary study, not proof.
The readings
key figures from the evidencestudy sample size
Why this verdict
Evidence
The cited study is real, the authors are correctly named, and the protocol described in the post matches it precisely. Doherty et al. (2023), published in Nutrients, asked participants to consume two medium-sized green kiwifruit (Actinidia deliciosa) one hour before bed for four weeks, following a one-week baseline period.
The reported results: total sleep time increased from a baseline of 7.6 ± 0.75 hours to 8.55 ± 0.44 hours at week 4, described by the authors as a statistically significant increase of 0.83 ± 0.23 hours (p < 0.05). Sleep efficiency increased from 86.2% at baseline to 91.5% (week 2), 93.4% (week 3), 93% (week 4) and 93.3% (week 5). Wake after sleep onset (WASO) reduced significantly across the intervention (Friedman's χ²(4) = 12.5, p < 0.05), with significant reductions at weeks 3, 4 and 5. Number of awakenings also fell significantly. PSQI global scores improved from 6.47 ± 2.17 at baseline to 4.13 ± 1.19 post-intervention (p = 0.002).
The critical omitted detail: the study population was 15 elite athletes (9 sailors, 6 middle-distance runners), not general poor sleepers. The paper's own limitations section states the sample size was small (n = 15). The design was open-label and single-arm, meaning every participant knew they were eating kiwifruit and there was no placebo or control group. Sleep was measured by the Consensus Sleep Diary, a self-report instrument, not by actigraphy or polysomnography. A peer-reviewed narrative review notes the study was conducted during the COVID-19 pandemic and that lockdown restrictions prevented the use of objective sleep measurements.
Findings
✓ What's accurate 7
- The study exists, is peer-reviewed, and the five authors are cited correctly
- The 2023 date is correct
- The protocol description is exact: two kiwifruit, one hour before bed, four weeks
- Total sleep time did increase substantially, from 7.6 h to 8.55 h. The "+1 hour" figure is a rounded-up version of a real finding
- WASO did decrease significantly, and sleep efficiency rose from 86% to 93%
- Self-reported sleep quality did improve significantly on the PSQI
- Kiwifruit does contain serotonin and antioxidants. The paper cites serotonin content at 5.8 µg/g and melatonin at 24 µg/g
≈ What's misleading 7
- Subgroup generalisation: the post frames this as advice for anyone "struggling to sleep." The study was in 15 elite national-level athletes with a normal baseline sleep duration of 7.6 hours. Nothing in it establishes what happens to a general adult with insomnia
- Omitted qualifier (most significant): no mention that the study had no control group and no placebo. In an open-label study where participants know they are receiving a "sleep intervention" and record their own sleep in a diary, expectancy effects alone can produce improvements of this size. The authors and subsequent reviewers both flag this
- Omitted qualifier: the sample was 15 people. The paper itself lists this as a limitation
- Omitted qualifier: all outcomes were self-reported. No actigraphy or polysomnography was used, because the study ran during COVID lockdowns
- Exaggeration (minor): "+1 hour" rounds up. The authors report the statistically significant increase as 0.83 hours, about 50 minutes. The raw baseline-to-week-4 difference is 0.95 hours
- Unsupported causal inference: "led to" implies causation. A single-arm before-after study cannot separate the kiwifruit from time, seasonal training changes, or the act of being observed
- Marketing-style mechanism framing: attributing the effect to "serotonin and antioxidants" states a hypothesis as if it were a demonstrated mechanism. The study measured no serotonin or melatonin levels. It offered these as plausible explanations only
? What's uncertain 3
- The specific "47%" WASO figure could not be located verbatim in the paper's text through the sources I could retrieve. The paper reports the statistical test result and significance for WASO but the percentage change appears to be a calculation derived from the study's figures, circulated by a secondary commentator. It is plausible and consistent with the direction and significance the paper reports, but I could not confirm the exact number against the primary text
- Whether the "1 hour" figure originated with the poster or was inherited from that same secondary summary. The identical pairing of "1 hour" and "47%" in a widely shared X post suggests the latter
- Whether the effect would replicate in non-athletes, in a controlled design, or with objective sleep measurement. No randomised placebo-controlled trial of kiwifruit for sleep of comparable length appears to exist
Sources
5 of 5 linked to recordsDoherty R, Madigan S, Nevill A, Warrington G, Ellis JG. "The Impact of Kiwifruit Consumption on the Sleep and Recovery of Elite Athletes." Nutrients. 2023;15(10):2274
"Nutrition Strategies to Promote Sleep in Elite Athletes: A Scoping Review." Sports. 2025;13(10):342
"Dietary Protocols to Promote and Improve Restful Sleep: A Narrative Review." Nutrition Reviews, 2025
Lin HH, Tsai PS, Fang SC, Liu JF. "Effect of kiwifruit consumption on sleep quality in adults with sleep problems." Asia Pac J Clin Nutr. 2011;20(2):169-174
X/Twitter post by William A. Wallace, PhD summarizing the study, containing the exact "47%" and "1 hour" figures in the same combination used by the post