Elmsword
← Sleep

Sleep · Research · 5 min read

Weighted blankets and insomnia: sleep-quality scores improved, insomnia severity did not differ

In a 1-month pilot randomized controlled trial of 102 adults with insomnia, a 5.5 kg weighted blanket improved PSQI sleep-quality scores more than a 3.5 kg normal blanket, but insomnia severity (ISI) and actigraphy-measured sleep did not differ significantly.

Weighted blankets spread deep pressure across the body, and occupational therapists increasingly recommend them for people with sleep disorders. A 2024 pilot randomized controlled trial in China tested one against an ordinary blanket in adults with insomnia. After 1 month, self-rated sleep quality on the Pittsburgh Sleep Quality Index (PSQI) improved more in the weighted blanket group. The trial's other primary outcome, the Insomnia Severity Index (ISI), did not differ significantly between groups, and neither did sleep measured by a device (actigraphy). It was a small, short trial, so the result is a first signal, not a settled answer. [1]

What the researchers did

This was a prospective pilot randomized controlled trial at three tertiary hospitals in Zhejiang, China (registration ChiCTR2300078011, registered retrospectively). Adults with clinical insomnia were referred from outpatient sleep clinics; everyone needed a PSQI score of 5 or more. People with active drug abuse, excessive use of sleeping pills, severe cognitive impairment or a history of major diseases such as coronary heart disease, cerebral infarction or cancer were excluded. [1]

In total, 102 participants were randomized 1:1. Their mean age was 44.0 years (range 19–78), 84 (82.4%) were women and 45 (44.1%) used sleep medicine. Of the 102, 95 (93.1%) completed the follow-up. [1]

  • Weighted blanket (n = 52): a 5.5 kg blanket sewn with glass beads. The authors cite earlier studies suggesting about 10% of body weight; the average participant weighed 56.3 kg.
  • Normal blanket (n = 50): a 3.5 kg blanket of identical shape, without glass beads.
  • Both groups: used the blanket for 1 month, with an actigraphy device (a smart monitoring belt placed under the bed sheet at chest level) during the first week.

Participants were told only that two blankets of different weights and identical shapes would be issued. The analysis was intention-to-treat. PSQI scores range from 0 to 21 and ISI scores from 0 to 28; on both, a higher score means worse sleep. [1]

Key findings

Weighted blanket vs normal blanket after 1 month. Questionnaire outcomes compared with generalized linear mixed models adjusted for age and sex; actigraphy changes compared over the first week. P < 0.05 was considered statistically significant. [1]
OutcomeResultP
PSQI sleep quality (primary)Mean change −4.1 points (weighted) vs −2.0 points (normal)0.006
ISI insomnia severity (primary)No significant between-group difference0.233
Daytime sleepinessGreater improvement with the weighted blanket0.017
Stress, anxiety, fatigueGreater improvement with the weighted blanketall < 0.05
Bodily pain (SF-36)Greater improvement with the weighted blanket0.010
Depression (CESD-10)No significant difference0.126
Night awakenings (actigraphy, first week)Mean change −1.4 (weighted) vs +1.0 (normal)0.280
Other actigraphy sleep measuresNo differences observed—
Weighted blanket vs normal blanket in adults with insomnia102 adults with insomnia were randomized to a 5.5 kg weighted blanket (52) or a 3.5 kg normal blanket (50) for 1 month, with actigraphy in the first week. PSQI fell by a mean of 4.1 points with the weighted blanket and 2.0 points with the normal blanket (P = 0.006). ISI (P = 0.233), night awakenings (P = 0.280), other actigraphy measures and depression (P = 0.126) did not differ significantly.Weighted blanket vs normal blanket in adults with insomniaPilot randomized controlled trial, 102 adults, three hospitals in Zhejiang, ChinaA. Study design102 adultsclinical insomnia, 1:1 randomWeighted blanket, n = 525.5 kg, glass beadsNormal blanket, n = 503.5 kg, identical shapevsBoth groups: 1 month of blanket use · actigraphy (a monitoring belt under the sheet) in the first weekB. Mean change in PSQI score after 1 month (PSQI 0–21, lower = better sleep quality)Weighted blanket−4.1 pointsNormal blanket−2.0 pointsBetween-group difference: P = 0.006C. No significant between-group differenceISI (insomnia severity, the other primary outcome): P = 0.233Night awakenings, actigraphy, first week: −1.4 (weighted) vs +1.0 (normal), P = 0.280Other actigraphy sleep measures: no differences observedDepression (CESD-10): P = 0.126Small pilot trial; 82.4% women; blinding incomplete; partly funded by the monitoring-belt maker.Source: Yu et al., BMC Psychiatry 2024, Abstract and Results.
Source: Yu et al., BMC Psychiatry 2024, Abstract and Results. Bars show the mean change in PSQI score after 1 month; a drop means better self-rated sleep quality. [1]

Self-rated sleep quality improved more with the weighted blanket. The PSQI score fell by a mean of 4.1 points in the weighted blanket group and 2.0 points in the normal blanket group. At baseline, the mean PSQI was 13.6 in the weighted blanket group and 12.3 in the normal blanket group. [1]

“For primary outcomes, the weighted blanket intervention showed a significant beneficial effect on the PSQI.”

Yu et al. · Results · full text

Insomnia severity did not differ significantly. The ISI was a primary outcome too, and the between-group difference was not significant. The authors note that the PSQI covers several dimensions of sleep quality, while the ISI focuses on how severe insomnia is, how much it disturbs life and how satisfied people are with their sleep. They suggest 1 month may be too short for that to change. [1]

“Notably, this study found a significant improvement in PSQI scores, but not in ISI.”

Yu et al. · Discussion · full text

Some daytime measures moved; depression did not. Daytime sleepiness, stress, anxiety, fatigue and bodily pain improved more in the weighted blanket group. Depression scores (CESD-10) did not differ significantly. These were all questionnaire scores. [1]

Device-measured sleep did not differ significantly. In the first week, night awakenings went down in the weighted blanket group (−1.4) and up in the normal blanket group (+1.0), but the difference was not significant. [1]

“No differences in other objective sleep parameters were observed.”

Yu et al. · Results · full text

No severe adverse events were reported. Three participants (2 weighted, 1 normal) reported feeling cold or uncomfortable with the blankets or the actigraphy. The authors add a caution that goes beyond this trial's adults: parents should closely supervise children using a weighted blanket and avoid covering the face, and older adults with frailty, limited mobility or severe dementia who cannot remove the blanket from their head must, to avoid the risk of suffocation, use the lightest blanket possible after consulting a clinician. [1] Separately, the NIH's Safe to Sleep campaign and the American Academy of Pediatrics advise against using weighted blankets on babies.

What this doesn't tell us

The trial was small and short: 102 people, 1 month of blanket use, and device monitoring for only the first week. The authors call for larger and longer trials. [1]

“A limitation of our study was that the sleep monitoring by actigraphy only lasted for one week; and thus, the objective long-term changes in sleep could not be observed.”

Yu et al. · Strengths and limitations · full text

The authors say the trial was not powered for insomnia severity or device-measured sleep, so a non-significant result there does not show that the blanket has no effect. [1]

“Fifth, this study was not powered to detect significant effects on ISI and objective measures, and it remains to be proven whether benefits are maintained across insomnia severity and objective sleep measures.”

Yu et al. · Future perspectives · full text

Blinding was incomplete. A few participants may have worked out which blanket they had, several outcome assessors were not masked, and outcomes were largely self-reported questionnaires. The control group also got a blanket, so there was no no-blanket comparison. [1]

The sample was narrow: three hospitals in one Chinese province, 82.4% women, and people with major diseases were excluded. The trial was registered retrospectively. It was partly funded by Damian Health Tech (Hangzhou), the company that made the monitoring belt; the paper states the funders had no role in the study, and the authors declare no competing interests. [1]

What to take away

Opinion, not a study result: If I were weighing a weighted blanket for insomnia, I would see this trial as a reason to expect possibly better-rated sleep quality, not lower insomnia severity. It was one small 1-month pilot, so results may differ from person to person. If sleep problems persist, talk to a doctor.

Reference

  1. BMC Psychiatry 2024[1] Yu J, Du J, Yang Z, Chen W, Sun S, Gan M, Cai Y, Zhang L, Sun K, Xu J, Xu Q, Ke J, Zhang L, Zhu Y, Liu Z. Effect of weighted blankets on sleep quality among adults with insomnia: a pilot randomized controlled trial. BMC Psychiatry. 2024;24:765.
    DOI: 10.1186/s12888-024-06218-9 · Full text

This is a research explainer for general education, not medical advice.

If sleep problems persist, talk to a doctor about your options.