Skip to main content
Tessa NewsLink
Paste a health news link, or browse

Source study found

Story checked

Smartwatch can help screen for obstructive sleep apnea (opens in a new tab)

medicalxpress.com · 2026-10-13

Short answerEvidenceSource

Short answer

Mixed

Mixed.

2 claims go further than the study. One other point was not covered by the paper.

  • 2 supported
  • 2 overstated
  • 1 not covered

Checked against the study summary. The full text wasn't available, so some details couldn't be settled either way.

Share this check

Follow the evidence trail
1
2

NewsLink checks it

Mixed

Two of five claims overstate the study. Two of five check out. One claim the study doesn't address.

  • 2 supported
  • 2 overstated
  • 1 not covered
Open claim evidence
3
Then inspect each claim

Evidence layer

Claim by claim

Each claim gets a verdict. Expand it to see the evidence directly below.

5 claims in this story

Showing all 5 claimsChoose a verdict to focus the list.

Claim 1 of 5OverstatedThe Samsung Galaxy Watch correctly distinguished people with moderate-to-severe sleep apnea 94% of the time, with 94% sensitivity and 67% specificity at its default setting.View evidenceHide evidence

As stated94% accuracy, 94% sensitivity, 67% specificity

Why this verdict

The sensitivity and specificity values are supported at the default eAHI threshold: 94.1% sensitivity and 66.7% specificity. However, the story’s wording that the watch 'correctly distinguished' people '94% of the time' or had '94% accuracy' appears to convert the reported AUROC of 0.94 into an accuracy/percent-correct metric. The profile reports AUROC 0.94, not 94% classification accuracy at the default setting.

1

Study evidence

Galaxy Watch eAHI discrimination for moderate-to-severe OSA (PSG AHI ≥15) had AUROC 0.94 (95% CI 0.889–0.980).AUROC 0.94 (95% CI 0.889–0.980)

“In this novel prospective study, we rigorously evaluated the diagnostic performance of the Samsung Galaxy Watch for detecting moderate-to-severe OSA (apnea-hypopnea index, AHI ≥ 15 events/h) compared to in-lab PSG using the AHI”
Claim 2 of 5OverstatedThe study involved 147 adults with an average age of 54 years, including people with prior moderate-to-severe sleep apnea or a high pretest likelihood of having it, and compared Galaxy Watch 6 results with overnight sleep studies over multiple nights at home and in a sleep center.View evidenceHide evidence

As stated147 participants; average age 54; two nights in lab and three nights at home

Why this verdict

Several design details are supported: 152 adults were enrolled, 147 completed both in-lab PSG nights, participants had prior moderate-to-severe OSA or STOP-Bang ≥3, and the protocol included two in-lab PSG nights separated by at least three watch-only home nights. But the profile does not verify the average age of 54 or the specific 'Galaxy Watch 6' model at abstract depth. More importantly, the story wording suggests comparison with overnight sleep studies across both home and sleep-center nights, while the profile says the at-home nights were watch-only and lacked concurrent PSG reference.

1

Study evidence

Galaxy Watch eAHI discrimination for moderate-to-severe OSA (PSG AHI ≥15) had AUROC 0.94 (95% CI 0.889–0.980).AUROC 0.94 (95% CI 0.889–0.980)

“In this novel prospective study, we rigorously evaluated the diagnostic performance of the Samsung Galaxy Watch for detecting moderate-to-severe OSA (apnea-hypopnea index, AHI ≥ 15 events/h) compared to in-lab PSG using the AHI”
2

Study evidence

The study implemented a repeated-measures protocol with two in-lab PSG nights separated by ≥3 watch-only at-home nights to assess smartwatch performance in both controlled and real-world contexts within the same prospective study.

“...wore the smartwatch during two in-lab PSG nights...: an initial in-lab PSG night, followed by ≥ 3 watch-only at-home nights, and then a second in-lab PSG night.”
Claim 3 of 5Not coveredThe authors and investigators say consumer sleep wearables may serve as an effective screening tool, but they are not intended to diagnose obstructive sleep apnea.View evidenceHide evidence

Why this verdict

The screening-potential portion is supported by the abstract-level profile, which says the authors conclude consumer-grade wearables have substantial potential for scalable OSA screening. However, the supplied abstract profile does not contain the quoted or attributed statement that the devices are 'not diagnostic' or 'not intended to diagnose.' That may be a reasonable clinical caveat, but it is not verifiable from the supplied abstract-depth paper profile.

1

Study evidence

Galaxy Watch eAHI discrimination for moderate-to-severe OSA (PSG AHI ≥15) had AUROC 0.94 (95% CI 0.889–0.980).AUROC 0.94 (95% CI 0.889–0.980)

“In this novel prospective study, we rigorously evaluated the diagnostic performance of the Samsung Galaxy Watch for detecting moderate-to-severe OSA (apnea-hypopnea index, AHI ≥ 15 events/h) compared to in-lab PSG using the AHI”
Claim 4 of 5SupportedNew research suggests that a consumer smartwatch can help identify adults with moderate-to-severe obstructive sleep apnea.View evidenceHide evidence

Why this verdict

The abstract-level profile supports a hedged screening-style claim: the study prospectively validated Samsung Galaxy Watch estimated AHI against in-lab PSG for detecting moderate-to-severe OSA, with AUROC 0.94 and high sensitivity at the default threshold. The paper profile also says the authors conclude consumer-grade wearables have substantial potential for scalable OSA screening.

1

Study evidence

Galaxy Watch eAHI discrimination for moderate-to-severe OSA (PSG AHI ≥15) had AUROC 0.94 (95% CI 0.889–0.980).AUROC 0.94 (95% CI 0.889–0.980)

“In this novel prospective study, we rigorously evaluated the diagnostic performance of the Samsung Galaxy Watch for detecting moderate-to-severe OSA (apnea-hypopnea index, AHI ≥ 15 events/h) compared to in-lab PSG using the AHI”
Claim 5 of 5SupportedAmong high-risk individuals with high hypoxic burden, the device achieved 100% sensitivity and specificity.View evidenceHide evidence

As stated100% sensitivity and 100% specificity

Why this verdict

The abstract-level profile reports that, for PSG-derived hypoxic-burden-defined high-risk OSA classification, the Galaxy Watch achieved 100% sensitivity and 100% specificity using the default eAHI threshold. The profile also notes important abstract-level caveats—such as missing subgroup size, confidence intervals, and HB cutpoint details—but the numerical claim itself is supported.

1

Study evidence

Within the PSG-derived hypoxic burden (HB) defined high-risk OSA group, the Galaxy Watch achieved 100% sensitivity and 100% specificity using the default eAHI threshold (reported in abstract).sensitivity 100%; specificity 100% (confidence intervals not reported in abstract)

“and 'high-risk' OSA subjects using the PSG-derived hypoxic burden (HB) criteria.”
Then look for missing context

Context layer

What the story left out

Important study details the story did not include.

  • Threshold-dependent performance: default eAHI threshold favored sensitivity, while the cohort-optimized threshold increased specificity to 94.9% and reduced sensitivity to 82.4%.

    The story reports the default-threshold numbers but does not convey the threshold trade-off or the cohort-optimized operating point, which is material to interpreting screening performance.

    From Prospective diagnostic accuracy

  • At-home nights were watch-only and did not have concurrent PSG reference; the abstract profile does not report separate home-night validation against PSG.

    The story’s wording implies that smartwatch results were compared with overnight sleep studies over multiple nights at home and in a sleep center. The profile instead says the PSG comparisons occurred during two in-lab nights, with intervening at-home watch-only nights.

    From prospective repeated-measures protocol

4 things the story did carry across
  • Primary diagnostic-accuracy validation of Samsung Galaxy Watch estimated AHI against in-lab PSG AHI for detecting moderate-to-severe OSA, with AUROC 0.94 and default-threshold sensitivity/specificity of 94.1%/66.7%.
  • Secondary hypoxic-burden analysis reported 100% sensitivity and 100% specificity for HB-defined high-risk versus low-risk OSA using the default threshold.
  • Study population was enriched for adults with prior moderate-to-severe OSA or high pretest likelihood, which may limit applicability to broad consumer screening populations.
  • The supplied abstract profile supports a screening-potential conclusion but does not itself verify a quoted author statement that the watch is 'not diagnostic.'
Then read the study layer

Study layer

Study at a glance

Scan the study first. Expand only the parts you want to inspect.

Pieces of work

3

Evidence read

study summary

Lead result

human in vivo

1Lead resulthuman in vivoProspectively validate Samsung Galaxy Watch estimated AHI (eAHI) against in-lab polysomnography (PSG) for detecting moderate-to-severe obstructive sleep apnea (OSA; PSG AHI ≥ 15).Prospective diagnostic accuracyExpand

In plain English

Prospective diagnostic accuracy study comparing Samsung Galaxy Watch estimated AHI (eAHI) to in-lab polysomnography (PSG) AHI for detecting moderate-to-severe OSA (PSG AHI ≥ 15). Enrolled 152 adults (age ≥22) with prior moderate-to-severe OSA or STOP‑Bang ≥3; 147 completed two in-lab PSG nights (total 1,850 h PSG sleep). The Galaxy Watch eAHI was evaluated using AUROC and threshold-based sensitivity/specificity (default eAHI threshold 15 and a cohort-optimized threshold 25.95); hypoxic burden (HB)-defined high-risk strata were also assessed. The device achieved AUROC 0.94 (95% CI 0.889–0.980) for detecting AHI ≥15. At the default eAHI threshold of 15, sensitivity was 94.1% (95% CI 84.1–98.0%) and specificity was 66.7% (95% CI 51.0–79.4%). Using a cohort-optimized eAHI threshold of 25.95 yielded sensitivity 82.4% (95% CI 69.7–90.4%) and specificity 94.9% (95% CI 83.1–98.6%). Within PSG-derived HB-defined high-risk OSA, the device achieved 100% sensitivity and 100% specificity using the default threshold. The authors conclude these results indicate substantial potential for consumer-grade wearables to detect moderate-to-severe OSA and to identify high hypoxic burden individuals.

Key findings

  • Galaxy Watch eAHI discrimination for moderate-to-severe OSA (PSG AHI ≥15) had AUROC 0.94 (95% CI 0.889–0.980).AUROC 0.94 (95% CI 0.889–0.980)
  • At the default eAHI threshold of 15, Galaxy Watch sensitivity for detecting AHI ≥15 was 94.1% (95% CI 84.1–98.0%) and specificity was 66.7% (95% CI 51.0–79.4%).Sensitivity 94.1% (95% CI 84.1–98.0%); Specificity 66.7% (95% CI 51.0–79.4%)
“In this novel prospective study, we rigorously evaluated the diagnostic performance of the Samsung Galaxy Watch for detecting moderate-to-severe OSA (apnea-hypopnea index, AHI ≥ 15 events/h) compared to in-lab PSG using the AHI”
2human in vivoEvaluate Galaxy Watch performance for identifying ‘high-risk’ OSA defined by PSG-derived hypoxic burden (HB) risk strata (low- vs high-risk), in addition to AHI-based severity.Prospective diagnostic/risk-stratification validationExpand

In plain English

Prospective, in-lab validation of the Samsung Galaxy Watch for classifying PSG-derived hypoxic burden (HB) risk strata (low vs. high). In the study cohort (152 enrolled; 147 completed two in-lab PSG nights; total 1,850 h PSG sleep), the abstract reports that the Galaxy Watch, using its default estimated AHI (eAHI) threshold, achieved 100% sensitivity and 100% specificity for discriminating HB-defined high-risk versus low-risk OSA.

Key findings

  • Within the PSG-derived hypoxic burden (HB) defined high-risk OSA group, the Galaxy Watch achieved 100% sensitivity and 100% specificity using the default eAHI threshold (reported in abstract).sensitivity 100%; specificity 100% (confidence intervals not reported in abstract)
“and 'high-risk' OSA subjects using the PSG-derived hypoxic burden (HB) criteria.”
What this piece can’t prove
  • Study population selection (prior moderate-to-severe OSA or STOP-Bang ≥ 3) may limit generalizability to broader screening populations.

2 further details could not be confirmed from the summary.

3human in vivoDemonstrate performance across controlled (in-lab PSG nights) and real-world (watch-only at-home nights) measurement context within the same prospective protocol.prospective repeated-measures protocolExpand

In plain English

The prospective protocol used a repeated-measures design in which 152 adults wore a consumer smartwatch during two in-lab polysomnography (PSG) nights separated by ≥3 watch-only at-home nights (147 participants completed both in-lab nights; total 1,850 h of in-lab PSG sleep). The design was framed as enabling assessment of device performance across controlled (in-lab PSG) and real-world (watch-only at-home) measurement contexts within the same study.

Key findings

  • The study implemented a repeated-measures protocol with two in-lab PSG nights separated by ≥3 watch-only at-home nights to assess smartwatch performance in both controlled and real-world contexts within the same prospective study.
  • High protocol completion for in-lab segments was reported: 147 of 152 participants completed both in-lab PSG nights (total 1,850 h of in-lab PSG sleep), supporting feasibility of the repeated in-lab measurements.
“...wore the smartwatch during two in-lab PSG nights...: an initial in-lab PSG night, followed by ≥ 3 watch-only at-home nights, and then a second in-lab PSG night.”
What this piece can’t prove
  • Abstract does not report separate analytic results comparing in-lab versus at-home watch-only performance against PSG.

2 further details could not be confirmed from the summary.

Finally, the search trail

Method layer

NewsLink found the paper. Tessa takes you deeper.

NewsLink checks the story. Tessa is where you inspect the paper, authors, evidence, and research context.

Papers considered

The selected paper, plus nearby candidates.

PubMed, Europe PMC, Crossref · 37 candidate papers

And 31 more candidates considered.