Digital delivery did not beat structured written exercises
A new randomized trial in PLOS One tested whether an internet-based vestibular rehabilitation program could reduce symptoms more than standardized written instructions after acute onset vertigo. The short answer is no: the digital tool was not superior at the 6-week primary endpoint.
That negative result is clinically useful rather than uninformative. This was not a trial of online rehabilitation against no rehabilitation. It was a pragmatic superiority trial against an active comparator: written instructions for the same vestibular exercises, plus advice intended to prevent inactivity, secondary dizziness, and fear of movement. For clinicians, the key message is therefore not that digital vestibular rehabilitation failed as a concept, but that a structured paper-based exercise pathway performed similarly under trial conditions.
Source boundary
Source boundary: The trial design, endpoints, effect estimates, adherence, safety findings, funding notes, and limitations discussed below come from the full-text PDF and associated PMC record for Surano et al., PLOS One, published June 12, 2026. Brief references to the clinical importance of acute vestibular syndrome, posterior-circulation stroke as an important differential diagnosis, and implementation barriers for supervised rehabilitation are external clinical context or background discussed by the authors; they should not be read as new outcome findings from this trial.
What the investigators studied
The investigators conducted a randomized, controlled, evaluator-blinded, multicentre superiority trial at nine hospitals in Sweden. Adults with ongoing acute vestibular syndrome were enrolled within 1-7 days after symptom onset. Of 184 randomized participants, 183 were included in analyses: 94 in the online vestibular rehabilitation arm and 89 in the written-instruction arm after one written-instruction participant was excluded for not meeting acute vestibular syndrome criteria.
The cohort was clinically recognizable for hospital-based acute vestibular care: median age was 56 years, approximately half of participants were female, and about 90% had a baseline diagnosis of vestibular neuritis. The trial was registered as NCT05056324, and the article reports PMID 42284306 and DOI 10.1371/journal.pone.0351092.
Intervention and comparator
The digital intervention was a 6-week web-based vestibular rehabilitation program. It provided personalized and progressively adjusted home exercises, progress tracking, reminders, and text/video instructions. The written-instruction arm received the same six home-based vestibular rehabilitation exercises and advice, with instructions to increase difficulty when possible. After 3 months, participants in the written-instruction group were offered access to the online tool.
This comparator choice matters. A weak control group can make a digital intervention look better than it would in routine practice. Here, the control was an active rehabilitation format, not a waiting list or generic usual care. That design makes the negative superiority result more credible for real implementation decisions.
Primary endpoint and effect estimate
The primary outcome was the between-group difference in vestibular symptoms at 6 weeks, measured using the Vertigo Symptom Scale Short Form (VSS-SF; range 0-60). The investigators prespecified that a difference of at least 3 points would be clinically significant.
Both groups improved. In the intention-to-treat analysis, mean VSS-SF fell to 11.1 in the online group and 13.1 in the written-instruction group. After adjustment for baseline symptom severity and prespecified covariates, adjusted mean VSS-SF was 12.2 for online rehabilitation and 14.1 for written instructions. The adjusted mean difference was -2.0 points, with a 95% confidence interval from -4.9 to 0.9 and p=0.18.
The per-protocol analysis pointed in the same direction but did not change the interpretation: adjusted mean difference -1.7 points, 95% CI -4.7 to 1.3, p=0.27. The observed difference was smaller than the prespecified 3-point clinical relevance threshold and statistically non-significant. The confidence interval leaves some uncertainty about a modest benefit, but the trial does not support a claim of digital superiority.
Secondary outcomes, adherence, and safety
Secondary outcomes also favored a cautious reading. Dizziness Handicap Inventory scores, balance testing, and walking speed improved over time in both groups, with no significant between-group differences. At 12 weeks, adjusted VSS-SF was 6.7 in the online group and 4.7 in the written-instruction group; the adjusted difference was 2.1 points, 95% CI -0.8 to 5.0, p=0.16. The authors also noted a small VSS-A autonomic-anxiety subscale difference at 12 weeks favoring written instructions, but it did not reach the threshold for clinical relevance and was interpreted cautiously in the setting of multiple comparisons.
Adherence was high in both arms. Using the trial definition of performing exercises at least once daily, 89-95% of participants were compliant during weeks 1-2, 69-81% during weeks 3-4, and 64-69% during weeks 5-6. No significant adherence differences were observed between study arms. This is important because the web tool’s reminders, videos, and personalization might be expected to improve engagement, yet the written instructions performed well when delivered within a structured trial.
The safety signal was reassuring. The trial reported 21 serious adverse events, including one death from myocardial infarction in the online group and one intensive care unit admission due to an unspecified cardiac event in the written-instruction group, but no serious adverse events were considered related to vestibular exercises. Two non-serious adverse events were considered likely exercise-related: transient arm numbness in one online-group participant and neck/shoulder pain in one written-instruction participant. Both discontinued exercises but completed follow-up.
Why a negative superiority trial still matters
For physician readers, the central interpretation is that delivery format may be less important than reliable access to a structured vestibular exercise program within this trial context. The trial does not show that the online program is ineffective; rather, it shows that it did not outperform well-constructed written instructions in this population and setting.
Several design features help explain the result. Participants were enrolled early after symptom onset, when spontaneous recovery and vestibular compensation may already be active. The written comparator contained the same core exercises rather than passive advice. Follow-up was frequent enough that both groups may have had higher engagement than would be expected in routine care. Under those conditions, the incremental value of a digital interface may be small.
This finding is especially relevant for health systems considering digital vestibular rehabilitation as a scalability strategy. The study supports flexibility in delivery format, but not preferential adoption on the basis of superior symptom reduction. A web-based tool may still be useful where written materials are poorly implemented, where reminders and video instruction improve usability, or where supervised physiotherapy is unavailable. Those are implementation hypotheses, not superiority claims proven by this endpoint.
Limitations and overclaim control
- The trial stopped after 184 randomized participants because recruitment was slower than expected; an updated power calculation was used.
- The anticipated effect size was derived from chronic vestibular-disorder data in primary care, whereas treatment effects in acute vestibular syndrome may be smaller.
- High adherence and follow-up within a trial may not reflect routine clinical practice.
- The cohort may represent a healthier and more educated subset of patients able to participate in exercise-based rehabilitation and trial procedures.
- Digital literacy and motivation were not systematically assessed; the PDF notes exclusions related to lack of digital access/proficiency and Swedish-language understanding.
- Most participants had vestibular neuritis, so the findings should not be generalized broadly to benign paroxysmal positional vertigo, Meniere disease, or stroke-related vertigo.
- The comparator was active written rehabilitation, so the trial does not test online rehabilitation against no rehabilitation.
Clinical implications
The practical takeaway is measured: in adults with acute vestibular syndrome treated in Swedish hospital settings, internet-based vestibular rehabilitation was not superior to structured written instructions for reducing vestibular symptoms at 6 weeks. Both groups improved, adherence was high, and no serious exercise-related adverse events were identified.
For clinicians, the priority is to ensure that appropriate patients receive clear vestibular rehabilitation exercises, understand progression, and remain engaged. Written instructions may be sufficient for many patients when they are standardized and actionable. Digital tools may be reasonable alternatives for patients who prefer app-like guidance, need video demonstrations, or have limited access to supervised rehabilitation, but the present PDF does not justify claiming better symptom outcomes than well-designed written instructions.
Acute vertigo still requires diagnostic discipline. External clinical context, also reflected in the article’s background, is that posterior-circulation stroke remains an important differential diagnosis in acute vestibular syndrome. Rehabilitation delivery should not be framed as a substitute for appropriate clinical evaluation, risk assessment, and follow-up.
Bottom line
This PDF deep read changes the emphasis from technology superiority to rehabilitation access. The digital program did not beat active written instructions on the prespecified 6-week VSS-SF endpoint. The study is most useful as evidence that structured vestibular exercises can be delivered through more than one format, and that implementation choices should be guided by patient preference, access, usability, and local care pathways rather than by an assumed digital advantage.
Source
Surano S, Lindell E, Mathé J, Davidsson H, Tomanovic T, Bjurman M, et al. Internet-based vestibular rehabilitation versus written instructions after acute vertigo: A randomised controlled trial. PLOS One. Published June 12, 2026. PMID: 42284306. PMCID: PMC13262863. DOI: 10.1371/journal.pone.0351092. Trial registration: NCT05056324.
